{"id":13934759,"url":"https://github.com/shaohua0116/ICLR2019-OpenReviewData","last_synced_at":"2025-07-19T19:31:44.817Z","repository":{"id":129631001,"uuid":"156375242","full_name":"shaohua0116/ICLR2019-OpenReviewData","owner":"shaohua0116","description":"Script that crawls meta data from ICLR OpenReview webpage. Tutorials on installing and using Selenium and ChromeDriver on Ubuntu.","archived":false,"fork":false,"pushed_at":"2019-12-05T19:33:11.000Z","size":57003,"stargazers_count":387,"open_issues_count":1,"forks_count":32,"subscribers_count":9,"default_branch":"master","last_synced_at":"2024-11-27T02:35:20.964Z","etag":null,"topics":["crawler","crawling-python","openreview","tutorial"],"latest_commit_sha":null,"homepage":"","language":"Jupyter Notebook","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/shaohua0116.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null}},"created_at":"2018-11-06T11:44:59.000Z","updated_at":"2024-10-16T07:05:15.000Z","dependencies_parsed_at":null,"dependency_job_id":"f3e98bda-7e50-4d87-a6b5-b46828d285c6","html_url":"https://github.com/shaohua0116/ICLR2019-OpenReviewData","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/shaohua0116/ICLR2019-OpenReviewData","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/shaohua0116%2FICLR2019-OpenReviewData","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/shaohua0116%2FICLR2019-OpenReviewData/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/shaohua0116%2FICLR2019-OpenReviewData/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/shaohua0116%2FICLR2019-OpenReviewData/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/shaohua0116","download_url":"https://codeload.github.com/shaohua0116/ICLR2019-OpenReviewData/tar.gz/refs/heads/master","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/shaohua0116%2FICLR2019-OpenReviewData/sbom","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":265998703,"owners_count":23862161,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2022-07-04T15:15:14.044Z","host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["crawler","crawling-python","openreview","tutorial"],"created_at":"2024-08-07T23:01:13.046Z","updated_at":"2025-07-19T19:31:42.546Z","avatar_url":"https://github.com/shaohua0116.png","language":"Jupyter Notebook","funding_links":[],"categories":["Jupyter Notebook"],"sub_categories":[],"readme":"# Crawl and Visualize ICLR 2019 OpenReview Data\n\n\u003cp align=\"center\"\u003e\n    \u003cimg src=\"asset/logo_wordcloud.png\" width=\"720\"/\u003e\n\u003c/p\u003e\n\n## Descriptions\n\nThis Jupyter Notebook contains the data and visualizations that are crawled ICLR 2019 OpenReview webpages. All the crawled data (sorted by the average ratings) can be found [here](#Data). The accepted papers have an average rating of 6.611 and 4.716 for rejected papers. The distributions are plotted as follows. \n\n\u003cp align=\"center\"\u003e\n    \u003cimg src=\"asset/decision.jpg\" width=\"800\"/\u003e\n\u003c/p\u003e\n\n## Prerequisites\n\n- Python3.5\n- [selenium](https://selenium-python.readthedocs.io/)\n- [pyvirtualdisplay](https://pypi.org/project/PyVirtualDisplay/) (run on a headless device)\n- [wordcloud](https://pypi.org/project/wordcloud/)\n- [imageio](https://imageio.github.io/)\n\n## Visualizations \n\n\nThe word clouds formed by keywords of submissions show the hot topics including **reinforcement learning**, **generative adversarial networks**, **generative models**, **imitation learning**, **representation learning**, etc.\n\u003cp align=\"center\"\u003e\n    \u003cimg src=\"asset/wordcloud.png\" width=\"720\"/\u003e\n\u003c/p\u003e\n\nThis figure is plotted with python [word cloud generator](https://github.com/amueller/word_cloud) \n\n```python\nfrom wordcloud import WordCloud\nwordcloud = WordCloud(max_font_size=64, max_words=160, \n                      width=1280, height=640,\n                      background_color=\"black\").generate(' '.join(keywords))\nplt.figure(figsize=(16, 8))\nplt.imshow(wordcloud, interpolation=\"bilinear\")\nplt.axis(\"off\")\nplt.show()\n```\n\nThe distributions of reviewer ratings center around 5 to 6 (mean: 5.15).\n\n\u003cp align=\"center\"\u003e\n    \u003cimg src=\"asset/rating.png\" width=\"640\"/\u003e\n\u003c/p\u003e\n\nYou can compute how many papers are beaten by yours with\n\n```python\ndef PR(rating_mean, your_rating):\n    pr = np.sum(your_rating \u003e= np.array(rating_mean))/len(rating_mean)*100\n    return pr\nmy_rating = (5+6+7)/3  # your average rating here\nprint('Your papar beats {:.2f}% of submission '\n      '(well, jsut based on the ratings...)'.format(PR(rating_mean, my_rating)))\n# ICLR 2017: accept rate 39.1% (198/507) (15 orals and 183 posters)\n# ICLR 2018: accept rate 32% (314/981) (23 orals and 291 posters)\n# ICLR 2018: accept rate ?% (?/1580)\n```\n\nThe top 50 common keywords and their frequency.\n\n\u003cp align=\"center\"\u003e\n    \u003cimg src=\"asset/frequency.png\" width=\"640\"/\u003e\n\u003c/p\u003e\n\nThe average reviewer ratings and the frequency of keywords indicate that to maximize your chance to get higher ratings would be using the keywords such as **theory**, **robustness**, or **graph neural network**.\n\n\u003cp align=\"center\"\u003e\n    \u003cimg src=\"asset/rating_frequency.png\" width=\"800\"/\u003e\n\u003c/p\u003e\n\n## How it works\n\nSee [How to install Selenium and ChromeDriver on Ubuntu](#Install).\n\nTo crawl data from dynamic websites such as OpenReview, a headless web simulator is created by\n\n```\nfrom selenium import webdriver\nfrom selenium.webdriver.chrome.options import Options\nexecutable_path = '/Users/waltersun/Desktop/chromedriver'  # path to your executable browser\noptions = Options()\noptions.add_argument(\"--headless\")\nbrowser = webdriver.Chrome(options=options, executable_path=executable_path)  \n```\n\nThen, we can get the content of a webpage\n\n```\nbrowser.get(url)\n```\n\nTo know what content we can crawl, we will need to inspect the webpage layout.\n\n\u003cp align=\"center\"\u003e\n    \u003cimg src=\"asset/inspect.png\" width=\"720\"/\u003e\n\u003c/p\u003e\n\nI chose to get the content by\n\n```\nkey = browser.find_elements_by_class_name(\"note_content_field\")\nvalue = browser.find_elements_by_class_name(\"note_content_value\")\n```\n\nThe data includes the abstract, keywords, TL; DR, comments.\n\n## \u003ca id=\"Install\"\u003e\u003c/a\u003eInstalling Selenium and ChromeDriver on Ubuntu\nThe following content is hugely borrowed from a nice [post](https://christopher.su/2015/selenium-chromedriver-ubuntu/) written by Christopher Su.\n\n- Install Google Chrome for Debian/Ubuntu\n\n```\nsudo apt-get install libxss1 libappindicator1 libindicator7\nwget https://dl.google.com/linux/direct/google-chrome-stable_current_amd64.deb\n\nsudo dpkg -i google-chrome*.deb\nsudo apt-get install -f\n```\n\n- Install `xvfb` to run Chrome on a headless device\n\n```\nsudo apt-get install xvfb\n```\n\n- Install ChromeDriver for 64-bit Linux\n\n```\nsudo apt-get install unzip  # If you don't have unzip package\n\nwget -N http://chromedriver.storage.googleapis.com/2.26/chromedriver_linux64.zip\nunzip chromedriver_linux64.zip\nchmod +x chromedriver\n\nsudo mv -f chromedriver /usr/local/share/chromedriver\nsudo ln -s /usr/local/share/chromedriver /usr/local/bin/chromedriver\nsudo ln -s /usr/local/share/chromedriver /usr/bin/chromedriver\n```\n\nIf your system is 32-bit, please find the ChromeDriver releases [here](http://chromedriver.storage.googleapis.com/) and modify the above download command.\n\n- Install Python dependencies (Selenium and pyvirtualdisplay)\n\n```\npip install pyvirtualdisplay selenium\n```\n\n- Test your setup in Python\n\n```python\nfrom pyvirtualdisplay import Display\nfrom selenium import webdriver\n\ndisplay = Display(visible=0, size=(1024, 1024))\ndisplay.start()\nbrowser = webdriver.Chrome()\nbrowser.get('http://shaohua0116.github.io/')\nprint(browser.title)\nprint(browser.find_element_by_class_name('bio').text)\n```\n\n## \u003ca id=\"Data\"\u003e\u003c/a\u003eAll ICLR 2019 OpenReview data\nCollected at 2019-12-05 11:31:13.692315\n\nNumber of submissions: 1579 (withdrawn submissions: 0)\n\n| Rank | Average Rating | Title | Ratings | Variance | Decision |\n| --- | --- | --- | --- | --- | --- |\n| 1 | 8.67 | [Generating High Fidelity Images With Subscale Pixel Networks And Multidimensional Upscaling](https://openreview.net/forum?id=HylzTiC5Km) | 7, 10, 9 | 1.25 | Accept (Oral) |\n| 2 | 8.67 | [Alista: Analytic Weights Are As Good As Learned Weights In Lista](https://openreview.net/forum?id=B1lnzn0ctQ) | 10, 7, 9 | 1.25 | Accept (Poster) |\n| 3 | 8.33 | [Benchmarking Neural Network Robustness To Common Corruptions And Perturbations](https://openreview.net/forum?id=HJz6tiCqYm) | 7, 9, 9 | 0.94 | Accept (Poster) |\n| 4 | 8.33 | [On Random Deep Weight-tied Autoencoders: Exact Asymptotic Analysis, Phase Transitions, And Implications To Training](https://openreview.net/forum?id=HJx54i05tX) | 9, 8, 8 | 0.47 | Accept (Oral) |\n| 5 | 8.00 | [Posterior Attention Models For Sequence To Sequence Learning](https://openreview.net/forum?id=BkltNhC9FX) | 8, 9, 7 | 0.82 | Accept (Poster) |\n| 6 | 8.00 | [Pay Less Attention With Lightweight And Dynamic Convolutions](https://openreview.net/forum?id=SkVhlh09tX) | 8, 8, 8 | 0.00 | Accept (Oral) |\n| 7 | 8.00 | [Slimmable Neural Networks](https://openreview.net/forum?id=H1gMCsAqY7) | 8, 9, 7 | 0.82 | Accept (Poster) |\n| 8 | 8.00 | [A Unified Theory Of Early Visual Representations From Retina To Cortex Through Anatomically Constrained Deep Cnns](https://openreview.net/forum?id=S1xq3oR5tQ) | 8, 8, 8 | 0.00 | Accept (Oral) |\n| 9 | 8.00 | [Ordered Neurons: Integrating Tree Structures Into Recurrent Neural Networks](https://openreview.net/forum?id=B1l6qiR5F7) | 9, 7, 8 | 0.82 | Accept (Oral) |\n| 10 | 8.00 | [Temporal Difference Variational Auto-encoder](https://openreview.net/forum?id=S1x4ghC9tQ) | 8, 9, 7 | 0.82 | Accept (Oral) |\n| 11 | 8.00 | [Enabling Factorized Piano Music Modeling And Generation With The Maestro Dataset](https://openreview.net/forum?id=r1lYRjC9F7) | 8, 8, 8 | 0.00 | Accept (Oral) |\n| 12 | 8.00 | [Variational Discriminator Bottleneck: Improving Imitation Learning, Inverse Rl, And Gans By Constraining Information Flow](https://openreview.net/forum?id=HyxPx3R9tm) | 6, 10, 8 | 1.63 | Accept (Poster) |\n| 13 | 8.00 | [Near-optimal Representation Learning For Hierarchical Reinforcement Learning](https://openreview.net/forum?id=H1emus0qF7) | 8, 9, 7 | 0.82 | Accept (Poster) |\n| 14 | 8.00 | [Ba-net: Dense Bundle Adjustment Networks](https://openreview.net/forum?id=B1gabhRcYX) | 9, 7, 8 | 0.82 | Accept (Oral) |\n| 15 | 8.00 | [Understanding And Improving Interpolation In Autoencoders Via An Adversarial Regularizer](https://openreview.net/forum?id=S1fQSiCcYm) | 7, 8, 9 | 0.82 | Accept (Poster) |\n| 16 | 8.00 | [Snip: Single-shot Network Pruning Based On Connection Sensitivity](https://openreview.net/forum?id=B1VZqjAcYX) | 8, 7, 9 | 0.82 | Accept (Poster) |\n| 17 | 8.00 | [Meta-learning Update Rules For Unsupervised Representation Learning](https://openreview.net/forum?id=HkNDsiC9KQ) | 8, 8, 8 | 0.00 | Accept (Oral) |\n| 18 | 8.00 | [Large Scale Gan Training For High Fidelity Natural Image Synthesis](https://openreview.net/forum?id=B1xsqj09Fm) | 8, 7, 9 | 0.82 | Accept (Oral) |\n| 19 | 8.00 | [Unsupervised Learning Of The Set Of Local Maxima](https://openreview.net/forum?id=H1lqZhRcFm) | 8, 8, 8 | 0.00 | Accept (Poster) |\n| 20 | 8.00 | [An Empirical Study Of Example Forgetting During Deep Neural Network Learning](https://openreview.net/forum?id=BJlxm30cKm) | 9, 8, 7 | 0.82 | Accept (Poster) |\n| 21 | 7.67 | [Learning Robust Representations By Projecting Superficial Statistics Out](https://openreview.net/forum?id=rJEjjoR9K7) | 7, 7, 9 | 0.94 | Accept (Oral) |\n| 22 | 7.67 | [Automatically Composing Representation Transformations As A Means For Generalization](https://openreview.net/forum?id=B1ffQnRcKX) | 7, 9, 7 | 0.94 | Accept (Poster) |\n| 23 | 7.67 | [Identifying And Controlling Important Neurons In Neural Machine Translation](https://openreview.net/forum?id=H1z-PsR5KX) | 7, 10, 6 | 1.70 | Accept (Poster) |\n| 24 | 7.67 | [Towards Robust, Locally Linear Deep Networks](https://openreview.net/forum?id=SylCrnCcFX) | 8, 8, 7 | 0.47 | Accept (Poster) |\n| 25 | 7.67 | [Deep Decoder: Concise Image Representations From Untrained Non-convolutional Networks](https://openreview.net/forum?id=rylV-2C9KQ) | 8, 8, 7 | 0.47 | Accept (Poster) |\n| 26 | 7.67 | [Lagging Inference Networks And Posterior Collapse In Variational Autoencoders](https://openreview.net/forum?id=rylDfnCqF7) | 7, 8, 8 | 0.47 | Accept (Poster) |\n| 27 | 7.67 | [A Variational Inequality Perspective On Generative Adversarial Networks](https://openreview.net/forum?id=r1laEnA5Ym) | 8, 8, 7 | 0.47 | Accept (Poster) |\n| 28 | 7.67 | [Robustness May Be At Odds With Accuracy](https://openreview.net/forum?id=SyxAb30cY7) | 8, 7, 8 | 0.47 | Accept (Poster) |\n| 29 | 7.67 | [Knockoffgan: Generating Knockoffs For Feature Selection Using Generative Adversarial Networks](https://openreview.net/forum?id=ByeZ5jC5YQ) | 6, 10, 7 | 1.70 | Accept (Oral) |\n| 30 | 7.67 | [Adaptive Input Representations For Neural Language Modeling](https://openreview.net/forum?id=ByxZX20qFQ) | 7, 8, 8 | 0.47 | Accept (Poster) |\n| 31 | 7.67 | [The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks](https://openreview.net/forum?id=rJl-b3RcF7) | 5, 9, 9 | 1.89 | Accept (Oral) |\n| 32 | 7.67 | [Critical Learning Periods In Deep Networks](https://openreview.net/forum?id=BkeStsCcKQ) | 9, 8, 6 | 1.25 | Accept (Poster) |\n| 33 | 7.67 | [Composing Complex Skills By Learning Transition Policies](https://openreview.net/forum?id=rygrBhC5tQ) | 7, 9, 7 | 0.94 | Accept (Poster) |\n| 34 | 7.67 | [Supervised Community Detection With Line Graph Neural Networks](https://openreview.net/forum?id=H1g0Z3A9Fm) | 6, 9, 8 | 1.25 | Accept (Poster) |\n| 35 | 7.67 | [Learning Deep Representations By Mutual Information Estimation And Maximization](https://openreview.net/forum?id=Bklr3j0cKX) | 7, 7, 9 | 0.94 | Accept (Oral) |\n| 36 | 7.67 | [Smoothing The Geometry Of Probabilistic Box Embeddings](https://openreview.net/forum?id=H1xSNiRcF7) | 8, 8, 7 | 0.47 | Accept (Oral) |\n| 37 | 7.67 | [A2bcd: Asynchronous Acceleration With Optimal Complexity](https://openreview.net/forum?id=rylIAsCqYm) | 7, 7, 9 | 0.94 | Accept (Poster) |\n| 38 | 7.67 | [Kernel Change-point Detection With Auxiliary Deep Generative Models](https://openreview.net/forum?id=r1GbfhRqF7) | 8, 8, 7 | 0.47 | Accept (Poster) |\n| 39 | 7.67 | [Imagenet-trained Cnns Are Biased Towards Texture; Increasing Shape Bias Improves Accuracy And Robustness](https://openreview.net/forum?id=Bygh9j09KX) | 7, 8, 8 | 0.47 | Accept (Oral) |\n| 40 | 7.67 | [Slalom: Fast, Verifiable And Private Execution Of Neural Networks In Trusted Hardware](https://openreview.net/forum?id=rJVorjCcKQ) | 7, 7, 9 | 0.94 | Accept (Oral) |\n| 41 | 7.67 | [Sparse Dictionary Learning By Dynamical Neural Networks](https://openreview.net/forum?id=B1gstsCqt7) | 6, 9, 8 | 1.25 | Accept (Poster) |\n| 42 | 7.50 | [On The Minimal Supervision For Training Any Binary Classifier From Only Unlabeled Data](https://openreview.net/forum?id=B1xWcj0qYm) | 7, 8, 8, 7 | 0.50 | Accept (Poster) |\n| 43 | 7.50 | [Exploration By Random Network Distillation](https://openreview.net/forum?id=H1lJJnR5Ym) | 4, 9, 10, 7 | 2.29 | Accept (Poster) |\n| 44 | 7.33 | [Dimensionality Reduction For Representing The Knowledge Of Probabilistic Models](https://openreview.net/forum?id=SygD-hCcF7) | 6, 7, 9 | 1.25 | Accept (Poster) |\n| 45 | 7.33 | [Probabilistic Recursive Reasoning For Multi-agent Reinforcement Learning](https://openreview.net/forum?id=rkl6As0cF7) | 8, 7, 7 | 0.47 | Accept (Poster) |\n| 46 | 7.33 | [Approximability Of Discriminators Implies Diversity In Gans](https://openreview.net/forum?id=rJfW5oA5KQ) | 8, 7, 7 | 0.47 | Accept (Poster) |\n| 47 | 7.33 | [Evaluating Robustness Of Neural Networks With Mixed Integer Programming](https://openreview.net/forum?id=HyGIdiRqtm) | 7, 8, 7 | 0.47 | Accept (Poster) |\n| 48 | 7.33 | [Biologically-plausible Learning Algorithms Can Scale To Large Datasets](https://openreview.net/forum?id=SygvZ209F7) | 9, 9, 4 | 2.36 | Accept (Poster) |\n| 49 | 7.33 | [Diagnosing And Enhancing Vae Models](https://openreview.net/forum?id=B1e0X3C9tQ) | 9, 6, 7 | 1.25 | Accept (Poster) |\n| 50 | 7.33 | [Learning To Navigate The Web](https://openreview.net/forum?id=BJemQ209FQ) | 7, 8, 7 | 0.47 | Accept (Poster) |\n| 51 | 7.33 | [Transferring Knowledge Across Learning Processes](https://openreview.net/forum?id=HygBZnRctX) | 6, 8, 8 | 0.94 | Accept (Oral) |\n| 52 | 7.33 | [Improving Differentiable Neural Computers Through Memory Masking, De-allocation, And Link Distribution Sharpness Control](https://openreview.net/forum?id=HyGEM3C9KQ) | 8, 7, 7 | 0.47 | Accept (Poster) |\n| 53 | 7.33 | [Towards Metamerism Via Foveated Style Transfer](https://openreview.net/forum?id=BJzbG20cFQ) | 7, 8, 7 | 0.47 | Accept (Poster) |\n| 54 | 7.33 | [Variance Reduction For Reinforcement Learning In Input-driven Environments](https://openreview.net/forum?id=Hyg1G2AqtQ) | 7, 9, 6 | 1.25 | Accept (Poster) |\n| 55 | 7.33 | [Quaternion Recurrent Neural Networks](https://openreview.net/forum?id=ByMHvs0cFQ) | 8, 7, 7 | 0.47 | Accept (Poster) |\n| 56 | 7.33 | [Promp: Proximal Meta-policy Search](https://openreview.net/forum?id=SkxXCi0qFX) | 6, 7, 9 | 1.25 | Accept (Poster) |\n| 57 | 7.33 | [Label Super-resolution Networks](https://openreview.net/forum?id=rkxwShA9Ym) | 7, 6, 9 | 1.25 | Accept (Poster) |\n| 58 | 7.33 | [Learning Self-imitating Diverse Policies](https://openreview.net/forum?id=HyxzRsR9Y7) | 8, 6, 8 | 0.94 | Accept (Poster) |\n| 59 | 7.33 | [Learning Protein Sequence Embeddings Using Information From Structure](https://openreview.net/forum?id=SygLehCqtm) | 7, 7, 8 | 0.47 | Accept (Poster) |\n| 60 | 7.33 | [Diffusion Scattering Transforms On Graphs](https://openreview.net/forum?id=BygqBiRcFQ) | 6, 9, 7 | 1.25 | Accept (Poster) |\n| 61 | 7.33 | [Deep Frank-wolfe For Neural Network Optimization](https://openreview.net/forum?id=SyVU6s05K7) | 7, 7, 8 | 0.47 | Accept (Poster) |\n| 62 | 7.33 | [Gradient Descent Aligns The Layers Of Deep Linear Networks](https://openreview.net/forum?id=HJflg30qKX) | 7, 9, 6 | 1.25 | Accept (Poster) |\n| 63 | 7.33 | [Recurrent Experience Replay In Distributed Reinforcement Learning](https://openreview.net/forum?id=r1lyTjAqYX) | 7, 7, 8 | 0.47 | Accept (Poster) |\n| 64 | 7.33 | [Large-scale Study Of Curiosity-driven Learning](https://openreview.net/forum?id=rJNwDjAqYX) | 6, 9, 7 | 1.25 | Accept (Poster) |\n| 65 | 7.33 | [Learning Localized Generative Models For 3d Point Clouds Via Graph Convolution](https://openreview.net/forum?id=SJeXSo09FQ) | 9, 6, 7 | 1.25 | Accept (Poster) |\n| 66 | 7.33 | [Prior Convictions: Black-box Adversarial Attacks With Bandits And Priors](https://openreview.net/forum?id=BkMiWhR5K7) | 7, 8, 7 | 0.47 | Accept (Poster) |\n| 67 | 7.33 | [Learning Latent Superstructures In Variational Autoencoders For Deep Multidimensional Clustering](https://openreview.net/forum?id=SJgNwi09Km) | 8, 7, 7 | 0.47 | Accept (Poster) |\n| 68 | 7.33 | [Learning Grid Cells As Vector Representation Of Self-position Coupled With Matrix Representation Of Self-motion](https://openreview.net/forum?id=Syx0Mh05YQ) | 8, 7, 7 | 0.47 | Accept (Poster) |\n| 69 | 7.33 | [Clarinet: Parallel Wave Generation In End-to-end Text-to-speech](https://openreview.net/forum?id=HklY120cYm) | 9, 6, 7 | 1.25 | Accept (Poster) |\n| 70 | 7.33 | [Dynamic Sparse Graph For Efficient Deep Learning](https://openreview.net/forum?id=H1goBoR9F7) | 8, 7, 7 | 0.47 | Accept (Poster) |\n| 71 | 7.33 | [Learning To Remember More With Less Memorization](https://openreview.net/forum?id=r1xlvi0qYm) | 7, 8, 7 | 0.47 | Accept (Oral) |\n| 72 | 7.33 | [Gan Dissection: Visualizing And Understanding Generative Adversarial Networks](https://openreview.net/forum?id=Hyg_X2C5FX) | 7, 7, 8 | 0.47 | Accept (Poster) |\n| 73 | 7.33 | [Detecting Egregious Responses In Neural Sequence-to-sequence Models](https://openreview.net/forum?id=HyNA5iRcFQ) | 7, 7, 8 | 0.47 | Accept (Poster) |\n| 74 | 7.33 | [Deep Layers As Stochastic Solvers](https://openreview.net/forum?id=ryxxCiRqYX) | 7, 7, 8 | 0.47 | Accept (Poster) |\n| 75 | 7.33 | [Small Nonlinearities In Activation Functions Create Bad Local Minima In Neural Networks](https://openreview.net/forum?id=rke_YiRct7) | 7, 7, 8 | 0.47 | Accept (Poster) |\n| 76 | 7.33 | [Efficient Training On Very Large Corpora Via Gramian Estimation](https://openreview.net/forum?id=Hke20iA9Y7) | 7, 8, 7 | 0.47 | Accept (Poster) |\n| 77 | 7.33 | [Diversity Is All You Need: Learning Skills Without A Reward Function](https://openreview.net/forum?id=SJx63jRqFm) | 8, 7, 7 | 0.47 | Accept (Poster) |\n| 78 | 7.33 | [Instagan: Instance-aware Image-to-image Translation](https://openreview.net/forum?id=ryxwJhC9YX) | 7, 8, 7 | 0.47 | Accept (Poster) |\n| 79 | 7.33 | [Time-agnostic Prediction: Predicting Predictable Video Frames](https://openreview.net/forum?id=SyzVb3CcFX) | 7, 8, 7 | 0.47 | Accept (Poster) |\n| 80 | 7.33 | [Learning To Schedule Communication In Multi-agent Reinforcement Learning](https://openreview.net/forum?id=SJxu5iR9KQ) | 7, 8, 7 | 0.47 | Accept (Poster) |\n| 81 | 7.33 | [No Training Required: Exploring Random Encoders For Sentence Classification](https://openreview.net/forum?id=BkgPajAcY7) | 7, 7, 8 | 0.47 | Accept (Poster) |\n| 82 | 7.33 | [Lanczosnet: Multi-scale Deep Graph Convolutional Networks](https://openreview.net/forum?id=BkedznAqKQ) | 7, 7, 8 | 0.47 | Accept (Poster) |\n| 83 | 7.33 | [The Neuro-symbolic Concept Learner: Interpreting Scenes, Words, And Sentences From Natural Supervision](https://openreview.net/forum?id=rJgMlhRctm) | 7, 6, 9 | 1.25 | Accept (Oral) |\n| 84 | 7.33 | [How Powerful Are Graph Neural Networks?](https://openreview.net/forum?id=ryGs6iA5Km) | 7, 7, 8 | 0.47 | Accept (Oral) |\n| 85 | 7.25 | [Episodic Curiosity Through Reachability](https://openreview.net/forum?id=SkeK3s0qKQ) | 7, 8, 6, 8 | 0.83 | Accept (Poster) |\n| 86 | 7.00 | [Strokenet: A Neural Painting Environment](https://openreview.net/forum?id=HJxwDiActX) | 7, 8, 6 | 0.82 | Accept (Poster) |\n| 87 | 7.00 | [Discriminator-actor-critic: Addressing Sample Inefficiency And Reward Bias In Adversarial Imitation Learning](https://openreview.net/forum?id=Hk4fpoA5Km) | 8, 6, 7 | 0.82 | Accept (Poster) |\n| 88 | 7.00 | [An Analytic Theory Of Generalization Dynamics And Transfer Learning In Deep Linear Networks](https://openreview.net/forum?id=ryfMLoCqtQ) | 8, 7, 6 | 0.82 | Accept (Poster) |\n| 89 | 7.00 | [Feature Intertwiner For Object Detection](https://openreview.net/forum?id=SyxZJn05YX) | 5, 9, 7 | 1.63 | Accept (Poster) |\n| 90 | 7.00 | [Learning Neural Pde Solvers With Convergence Guarantees](https://openreview.net/forum?id=rklaWn0qK7) | 7, 8, 6 | 0.82 | Accept (Poster) |\n| 91 | 7.00 | [Knowledge Flow: Improve Upon Your Teachers](https://openreview.net/forum?id=BJeOioA9Y7) | 6, 8, 7 | 0.82 | Accept (Poster) |\n| 92 | 7.00 | [Multilingual Neural Machine Translation With Knowledge Distillation](https://openreview.net/forum?id=S1gUsoR9YX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 93 | 7.00 | [Texttovec: Deep Contextualized Neural Autoregressive Topic Models Of Language With Distributed Compositional Prior](https://openreview.net/forum?id=rkgoyn09KQ) | 7, 8, 6 | 0.82 | Accept (Poster) |\n| 94 | 7.00 | [Supervised Policy Update For Deep Reinforcement Learning](https://openreview.net/forum?id=SJxTroR9F7) | 9, 6, 6 | 1.41 | Accept (Poster) |\n| 95 | 7.00 | [Gansynth: Adversarial Neural Audio Synthesis](https://openreview.net/forum?id=H1xQVn09FX) | 6, 7, 8 | 0.82 | Accept (Poster) |\n| 96 | 7.00 | [Lemonade: Learned Motif And Neuronal Assembly Detection In Calcium Imaging Videos](https://openreview.net/forum?id=SkloDjAqYm) | 8, 5, 8 | 1.41 | Accept (Poster) |\n| 97 | 7.00 | [The Comparative Power Of Relu Networks And Polynomial Kernels In The Presence Of Sparse Latent Structure](https://openreview.net/forum?id=rJgTTjA9tX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 98 | 7.00 | [Execution-guided Neural Program Synthesis](https://openreview.net/forum?id=H1gfOiAqYm) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 99 | 7.00 | [Deterministic Variational Inference For Robust Bayesian Neural Networks](https://openreview.net/forum?id=B1l08oAct7) | 7, 7, 7 | 0.00 | Accept (Oral) |\n| 100 | 7.00 | [Distributional Concavity Regularization For Gans](https://openreview.net/forum?id=SklEEnC5tQ) | 7, 8, 6, 7 | 0.71 | Accept (Poster) |\n| 101 | 7.00 | [Som-vae: Interpretable Discrete Representation Learning On Time Series](https://openreview.net/forum?id=rygjcsR9Y7) | 9, 6, 6 | 1.41 | Accept (Poster) |\n| 102 | 7.00 | [Variational Autoencoders With Jointly Optimized Latent Dependency Structure](https://openreview.net/forum?id=SJgsCjCqt7) | 7, 6, 8 | 0.82 | Accept (Poster) |\n| 103 | 7.00 | [Learning Sparse Relational Transition Models](https://openreview.net/forum?id=SJxsV2R5FQ) | 6, 7, 8 | 0.82 | Accept (Poster) |\n| 104 | 7.00 | [Adversarial Domain Adaptation For Stable Brain-machine Interfaces](https://openreview.net/forum?id=Hyx6Bi0qYm) | 9, 5, 7 | 1.63 | Accept (Poster) |\n| 105 | 7.00 | [The Role Of Over-parametrization In Generalization Of Neural Networks](https://openreview.net/forum?id=BygfghAcYX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 106 | 7.00 | [Differentiable Learning-to-normalize Via Switchable Normalization](https://openreview.net/forum?id=ryggIs0cYQ) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 107 | 7.00 | [Stochastic Optimization Of Sorting Networks Via Continuous Relaxations](https://openreview.net/forum?id=H1eSS3CcKX) | 8, 7, 6 | 0.82 | Accept (Poster) |\n| 108 | 7.00 | [A Statistical Approach To Assessing Neural Network Robustness](https://openreview.net/forum?id=S1xcx3C5FX) | 6, 7, 8 | 0.82 | Accept (Poster) |\n| 109 | 7.00 | [Darts: Differentiable Architecture Search](https://openreview.net/forum?id=S1eYHoC5FX) | 6, 7, 8 | 0.82 | Accept (Poster) |\n| 110 | 7.00 | [Learning Concise Representations For Regression By Evolving Networks Of Trees](https://openreview.net/forum?id=Hke-JhA9Y7) | 7, 6, 8 | 0.82 | Accept (Poster) |\n| 111 | 7.00 | [Padam: Closing The Generalization Gap Of Adaptive Gradient Methods In Training Deep Neural Networks](https://openreview.net/forum?id=BJll6o09tm) | 6, 6, 9 | 1.41 | Reject |\n| 112 | 7.00 | [A Universal Music Translation Network](https://openreview.net/forum?id=HJGkisCcKm) | 8, 7, 6 | 0.82 | Accept (Poster) |\n| 113 | 7.00 | [Deep Learning 3d Shapes Using Alt-az Anisotropic 2-sphere Convolution](https://openreview.net/forum?id=rkeSiiA5Fm) | 6, 8, 7 | 0.82 | Accept (Poster) |\n| 114 | 7.00 | [Energy-constrained Compression For Deep Neural Networks Via Weighted Sparse Projection And Layer Input Masking](https://openreview.net/forum?id=BylBr3C9K7) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 115 | 7.00 | [Deep Graph Infomax](https://openreview.net/forum?id=rklz9iAcKQ) | 9, 5, 7 | 1.63 | Accept (Poster) |\n| 116 | 7.00 | [On The Universal Approximability And Complexity Bounds Of Quantized Relu Neural Networks](https://openreview.net/forum?id=SJe9rh0cFX) | 7, 6, 8 | 0.82 | Accept (Poster) |\n| 117 | 7.00 | [Global-to-local Memory Pointer Networks For Task-oriented Dialogue](https://openreview.net/forum?id=ryxnHhRqFm) | 8, 8, 5 | 1.41 | Accept (Poster) |\n| 118 | 7.00 | [Self-monitoring Navigation Agent Via Auxiliary Progress Estimation](https://openreview.net/forum?id=r1GAsjC5Fm) | 8, 6, 7 | 0.82 | Accept (Poster) |\n| 119 | 7.00 | [Signsgd Via Zeroth-order Oracle](https://openreview.net/forum?id=BJe-DsC5Fm) | 8, 7, 6 | 0.82 | Accept (Poster) |\n| 120 | 7.00 | [Learning Particle Dynamics For Manipulating Rigid Bodies, Deformable Objects, And Fluids](https://openreview.net/forum?id=rJgbSn09Ym) | 8, 6, 7 | 0.82 | Accept (Poster) |\n| 121 | 7.00 | [Generative Code Modeling With Graphs](https://openreview.net/forum?id=Bke4KsA5FX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 122 | 7.00 | [The Deep Weight Prior](https://openreview.net/forum?id=ByGuynAct7) | 6, 8, 7 | 0.82 | Accept (Poster) |\n| 123 | 7.00 | [Bounce And Learn: Modeling Scene Dynamics With Real-world Bounces](https://openreview.net/forum?id=BJxssoA5KX) | 6, 7, 8 | 0.82 | Accept (Poster) |\n| 124 | 7.00 | [Quasi-hyperbolic Momentum And Adam For Deep Learning](https://openreview.net/forum?id=S1fUpoR5FQ) | 7, 6, 8 | 0.82 | Accept (Poster) |\n| 125 | 7.00 | [Integer Networks For Data Compression With Latent-variable Models](https://openreview.net/forum?id=S1zz2i0cY7) | 6, 7, 8 | 0.82 | Accept (Poster) |\n| 126 | 7.00 | [Deep Online Learning Via Meta-learning: Continual Adaptation For Model-based Rl](https://openreview.net/forum?id=HyxAfnA5tm) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 127 | 7.00 | [Are Adversarial Examples Inevitable?](https://openreview.net/forum?id=r1lWUoA9FQ) | 7, 8, 6 | 0.82 | Accept (Poster) |\n| 128 | 7.00 | [Learning To Screen For Fast Softmax Inference On Large Vocabulary Neural Networks](https://openreview.net/forum?id=ByeMB3Act7) | 7, 6, 8 | 0.82 | Accept (Poster) |\n| 129 | 7.00 | [Information-directed Exploration For Deep Reinforcement Learning](https://openreview.net/forum?id=Byx83s09Km) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 130 | 7.00 | [Rotdcf: Decomposition Of Convolutional Filters For Rotation-equivariant Deep Networks](https://openreview.net/forum?id=H1gTEj09FX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 131 | 7.00 | [Theoretical Analysis Of Auto Rate-tuning By Batch Normalization](https://openreview.net/forum?id=rkxQ-nA9FX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 132 | 7.00 | [Visual Semantic Navigation Using Scene Priors](https://openreview.net/forum?id=HJeRkh05Km) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 133 | 7.00 | [Woulda, Coulda, Shoulda: Counterfactually-guided Policy Search](https://openreview.net/forum?id=BJG0voC9YQ) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 134 | 7.00 | [Function Space Particle Optimization For Bayesian Neural Networks](https://openreview.net/forum?id=BkgtDsCcKQ) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 135 | 7.00 | [Eidetic 3d Lstm: A Model For Video Prediction And Beyond](https://openreview.net/forum?id=B1lKS2AqtX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 136 | 7.00 | [Wizard Of Wikipedia: Knowledge-powered Conversational Agents](https://openreview.net/forum?id=r1l73iRqKm) | 7, 6, 8 | 0.82 | Accept (Poster) |\n| 137 | 7.00 | [Meta-learning Probabilistic Inference For Prediction](https://openreview.net/forum?id=HkxStoC5F7) | 7, 6, 8 | 0.82 | Accept (Poster) |\n| 138 | 7.00 | [Don't Settle For Average, Go For The Max: Fuzzy Sets And Max-pooled Word Vectors](https://openreview.net/forum?id=SkxXg2C5FX) | 8, 8, 5 | 1.41 | Accept (Poster) |\n| 139 | 7.00 | [Solving The Rubik's Cube With Approximate Policy Iteration](https://openreview.net/forum?id=Hyfn2jCcKm) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 140 | 7.00 | [Learning A Meta-solver For Syntax-guided Program Synthesis](https://openreview.net/forum?id=Syl8Sn0cK7) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 141 | 7.00 | [Rotate: Knowledge Graph Embedding By Relational Rotation In Complex Space](https://openreview.net/forum?id=HkgEQnRqYQ) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 142 | 7.00 | [Generative Question Answering: Learning To Answer The Whole Question](https://openreview.net/forum?id=Bkx0RjA9tX) | 7, 6, 8 | 0.82 | Accept (Poster) |\n| 143 | 7.00 | [Local Sgd Converges Fast And Communicates Little](https://openreview.net/forum?id=S1g2JnRcFX) | 8, 5, 8 | 1.41 | Accept (Poster) |\n| 144 | 7.00 | [Ffjord: Free-form Continuous Dynamics For Scalable Reversible Generative Models](https://openreview.net/forum?id=rJxgknCcK7) | 7, 7, 7 | 0.00 | Accept (Oral) |\n| 145 | 7.00 | [Adashift: Decorrelation And Convergence Of Adaptive Learning Rate Methods](https://openreview.net/forum?id=HkgTkhRcKQ) | 6, 6, 9 | 1.41 | Accept (Poster) |\n| 146 | 7.00 | [What Do You Learn From Context? Probing For Sentence Structure In Contextualized Word Representations](https://openreview.net/forum?id=SJzSgnRcKX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 147 | 7.00 | [Modeling Uncertainty With Hedged Instance Embeddings](https://openreview.net/forum?id=r1xQQhAqKX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 148 | 7.00 | [Learning Implicitly Recurrent Cnns Through Parameter Sharing](https://openreview.net/forum?id=rJgYxn09Fm) | 8, 7, 6 | 0.82 | Accept (Poster) |\n| 149 | 7.00 | [Arm: Augment-reinforce-merge Gradient For Stochastic Binary Networks](https://openreview.net/forum?id=S1lg0jAcYm) | 8, 6, 7 | 0.82 | Accept (Poster) |\n| 150 | 7.00 | [On The Loss Landscape Of A Class Of Deep Neural Networks With No Bad Local Valleys](https://openreview.net/forum?id=HJgXsjA5tQ) | 7, 8, 6 | 0.82 | Accept (Poster) |\n| 151 | 7.00 | [Riemannian Adaptive Optimization Methods](https://openreview.net/forum?id=r1eiqi09K7) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 152 | 7.00 | [Learning To Learn Without Forgetting By Maximizing Transfer And Minimizing Interference](https://openreview.net/forum?id=B1gTShAct7) | 6, 8, 7 | 0.82 | Accept (Poster) |\n| 153 | 7.00 | [G-sgd: Optimizing Relu Neural Networks In Its Positively Scale-invariant Space](https://openreview.net/forum?id=SyxfEn09Y7) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 154 | 7.00 | [Reasoning About Physical Interactions With Object-oriented Prediction And Planning](https://openreview.net/forum?id=HJx9EhC9tQ) | 5, 7, 9 | 1.63 | Accept (Poster) |\n| 155 | 7.00 | [Hindsight Policy Gradients](https://openreview.net/forum?id=Bkg2viA5FQ) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 156 | 7.00 | [Unsupervised Domain Adaptation For Distance Metric Learning](https://openreview.net/forum?id=BklhAj09K7) | 8, 5, 8 | 1.41 | Accept (Poster) |\n| 157 | 7.00 | [Learning Mixed-curvature Representations In Product Spaces](https://openreview.net/forum?id=HJxeWnCcF7) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 158 | 7.00 | [Auxiliary Variational Mcmc](https://openreview.net/forum?id=r1NJqsRctX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 159 | 7.00 | [Unsupervised Speech Recognition Via Segmental Empirical Output Distribution Matching](https://openreview.net/forum?id=Bylmkh05KX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 160 | 7.00 | [On Computation And Generalization Of Generative Adversarial Networks Under Spectrum Control](https://openreview.net/forum?id=rJNH6sAqY7) | 8, 6, 7 | 0.82 | Accept (Poster) |\n| 161 | 7.00 | [Optimal Control Via Neural Networks: A Convex Approach](https://openreview.net/forum?id=H1MW72AcK7) | 6, 8, 7 | 0.82 | Accept (Poster) |\n| 162 | 7.00 | [Whitening And Coloring Batch Transform For Gans](https://openreview.net/forum?id=S1x2Fj0qKQ) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 163 | 7.00 | [Deep, Skinny Neural Networks Are Not Universal Approximators](https://openreview.net/forum?id=ryGgSsAcFQ) | 6, 8, 7 | 0.82 | Accept (Poster) |\n| 164 | 7.00 | [Nadpex: An On-policy Temporally Consistent Exploration Method For Deep Reinforcement Learning](https://openreview.net/forum?id=rkxciiC9tm) | 8, 6, 7 | 0.82 | Accept (Poster) |\n| 165 | 7.00 | [Learning To Solve Circuit-sat: An Unsupervised Differentiable Approach](https://openreview.net/forum?id=BJxgz2R9t7) | 6, 8, 7 | 0.82 | Accept (Poster) |\n| 166 | 7.00 | [A Convergence Analysis Of Gradient Descent For Deep Linear Neural Networks](https://openreview.net/forum?id=SkMQg3C5K7) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 167 | 7.00 | [Learning A Sat Solver From Single-bit Supervision](https://openreview.net/forum?id=HJMC_iA5tm) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 168 | 7.00 | [Generating Multiple Objects At Spatially Distinct Locations](https://openreview.net/forum?id=H1edIiA9KQ) | 6, 8, 7 | 0.82 | Accept (Poster) |\n| 169 | 7.00 | [K For The Price Of 1: Parameter-efficient Multi-task And Transfer Learning](https://openreview.net/forum?id=BJxvEh0cFQ) | 7, 6, 8 | 0.82 | Accept (Poster) |\n| 170 | 7.00 | [Bias-reduced Uncertainty Estimation For Deep Neural Classifiers](https://openreview.net/forum?id=SJfb5jCqKm) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 171 | 7.00 | [Probabilistic Neural-symbolic Models For Interpretable Visual Question Answering](https://openreview.net/forum?id=ryxhB3CcK7) | 8, 6, 7 | 0.82 | Reject |\n| 172 | 7.00 | [Representation Degeneration Problem In Training Natural Language Generation Models](https://openreview.net/forum?id=SkEYojRqtm) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 173 | 7.00 | [Neural Network Gradient-based Learning Of Black-box Function Interfaces](https://openreview.net/forum?id=r1e13s05YX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 174 | 7.00 | [A Data-driven And Distributed Approach To Sparse Signal Representation And Recovery](https://openreview.net/forum?id=B1xVTjCqKQ) | 8, 7, 6 | 0.82 | Accept (Poster) |\n| 175 | 7.00 | [Relaxed Quantization For Discretized Neural Networks](https://openreview.net/forum?id=HkxjYoCqKX) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 176 | 7.00 | [Invariant And Equivariant Graph Networks](https://openreview.net/forum?id=Syx72jC9tm) | 8, 4, 9 | 2.16 | Accept (Poster) |\n| 177 | 7.00 | [Dyrep: Learning Representations Over Dynamic Graphs](https://openreview.net/forum?id=HyePrhR5KX) | 6, 7, 8 | 0.82 | Accept (Poster) |\n| 178 | 7.00 | [The Laplacian In Rl: Learning Representations With Efficient Approximations](https://openreview.net/forum?id=HJlNpoA5YQ) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 179 | 7.00 | [Learning Recurrent Binary/ternary Weights](https://openreview.net/forum?id=HkNGYjR9FX) | 6, 8, 7 | 0.82 | Accept (Poster) |\n| 180 | 7.00 | [How Important Is A Neuron](https://openreview.net/forum?id=SylKoo0cKm) | 7, 7, 7 | 0.00 | Accept (Poster) |\n| 181 | 6.80 | [Subgradient Descent Learns Orthogonal Dictionaries](https://openreview.net/forum?id=HklSf3CqKm) | 7, 7, 7, 7, 6 | 0.40 | Accept (Poster) |\n| 182 | 6.75 | [Unsupervised Learning Via Meta-learning](https://openreview.net/forum?id=r1My6sR9tX) | 7, 6, 8, 6 | 0.83 | Accept (Poster) |\n| 183 | 6.75 | [Bayesian Deep Convolutional Networks With Many Channels Are Gaussian Processes](https://openreview.net/forum?id=B1g30j0qF7) | 7, 7, 7, 6 | 0.43 | Accept (Poster) |\n| 184 | 6.75 | [Deterministic Pac-bayesian Generalization Bounds For Deep Networks Via Generalizing Noise-resilience](https://openreview.net/forum?id=Hygn2o0qKX) | 8, 7, 7, 5 | 1.09 | Accept (Poster) |\n| 185 | 6.67 | [Structured Adversarial Attack: Towards General Implementation And Better Interpretability](https://openreview.net/forum?id=BkgzniCqY7) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 186 | 6.67 | [Adaptive Estimators Show Information Compression In Deep Neural Networks](https://openreview.net/forum?id=SkeZisA5t7) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 187 | 6.67 | [Tree-structured Recurrent Switching Linear Dynamical Systems For Multi-scale Modeling](https://openreview.net/forum?id=HkzRQhR9YX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 188 | 6.67 | [Residual Non-local Attention Networks For Image Restoration](https://openreview.net/forum?id=HkeGhoA5FX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 189 | 6.67 | [Cem-rl: Combining Evolutionary And Gradient-based Methods For Policy Search](https://openreview.net/forum?id=BkeU5j0ctQ) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 190 | 6.67 | [Marginal Policy Gradients: A Unified Family Of Estimators For Bounded Action Spaces With Applications](https://openreview.net/forum?id=HkgqFiAcFm) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 191 | 6.67 | [Relgan: Relational Generative Adversarial Networks For Text Generation](https://openreview.net/forum?id=rJedV3R5tm) | 6, 8, 6 | 0.94 | Accept (Poster) |\n| 192 | 6.67 | [Defensive Quantization: When Efficiency Meets Robustness](https://openreview.net/forum?id=ryetZ20ctX) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 193 | 6.67 | [Policy Transfer With Strategy Optimization](https://openreview.net/forum?id=H1g6osRcFQ) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 194 | 6.67 | [Big-little Net: An Efficient Multi-scale Feature Representation For Visual And Speech Recognition](https://openreview.net/forum?id=HJMHpjC9Ym) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 195 | 6.67 | [Universal Transformers](https://openreview.net/forum?id=HyzdRiR9Y7) | 6, 6, 8 | 0.94 | Accept (Poster) |\n| 196 | 6.67 | [Active Learning With Partial Feedback](https://openreview.net/forum?id=HJfSEnRqKQ) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 197 | 6.67 | [There Are Many Consistent Explanations Of Unlabeled Data: Why You Should Average](https://openreview.net/forum?id=rkgKBhA5Y7) | 6, 8, 6 | 0.94 | Accept (Poster) |\n| 198 | 6.67 | [Unsupervised Control Through Non-parametric Discriminative Rewards](https://openreview.net/forum?id=r1eVMnA9K7) | 8, 5, 7 | 1.25 | Accept (Poster) |\n| 199 | 6.67 | [On The Convergence Of A Class Of Adam-type Algorithms For Non-convex Optimization](https://openreview.net/forum?id=H1x-x309tm) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 200 | 6.67 | [Adaptivity Of Deep Relu Network For Learning In Besov And Mixed Smooth Besov Spaces: Optimal Rate And Curse Of Dimensionality](https://openreview.net/forum?id=H1ebTsActm) | 8, 6, 6 | 0.94 | Accept (Poster) |\n| 201 | 6.67 | [Predicting The Generalization Gap In Deep Networks With Margin Distributions](https://openreview.net/forum?id=HJlQfnCqKX) | 5, 9, 6 | 1.70 | Accept (Poster) |\n| 202 | 6.67 | [A Mean Field Theory Of Batch Normalization](https://openreview.net/forum?id=SyMDXnCcF7) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 203 | 6.67 | [Don't Let Your Discriminator Be Fooled](https://openreview.net/forum?id=HJE6X305Fm) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 204 | 6.67 | [L-shapley And C-shapley: Efficient Model Interpretation For Structured Data](https://openreview.net/forum?id=S1E3Ko09F7) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 205 | 6.67 | [Hyperbolic Attention Networks](https://openreview.net/forum?id=rJxHsjRqFQ) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 206 | 6.67 | [Learning To Make Analogies By Contrasting Abstract Relational Structure](https://openreview.net/forum?id=SylLYsCcFm) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 207 | 6.67 | [Meta-learning For Stochastic Gradient Mcmc](https://openreview.net/forum?id=HkeoOo09YX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 208 | 6.67 | [Directed-info Gail: Learning Hierarchical Policies From Unsegmented Demonstrations Using Directed Information](https://openreview.net/forum?id=BJeWUs05KQ) | 6, 6, 8 | 0.94 | Accept (Poster) |\n| 209 | 6.67 | [Building Dynamic Knowledge Graphs From Text Using Machine Reading Comprehension](https://openreview.net/forum?id=S1lhbnRqF7) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 210 | 6.67 | [Proxquant: Quantized Neural Networks Via Proximal Operators](https://openreview.net/forum?id=HyzMyhCcK7) | 8, 7, 5 | 1.25 | Accept (Poster) |\n| 211 | 6.67 | [Emergent Coordination Through Competition](https://openreview.net/forum?id=BkG8sjR5Km) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 212 | 6.67 | [Doubly Reparameterized Gradient Estimators For Monte Carlo Objectives](https://openreview.net/forum?id=HkG3e205K7) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 213 | 6.67 | [Learning To Understand Goal Specifications By Modelling Reward](https://openreview.net/forum?id=H1xsSjC9Ym) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 214 | 6.67 | [Off-policy Evaluation And Learning From Logged Bandit Feedback: Error Reduction Via Surrogate Policy](https://openreview.net/forum?id=HklKui0ct7) | 6, 8, 6 | 0.94 | Accept (Poster) |\n| 215 | 6.67 | [Improving Mmd-gan Training With Repulsive Loss Function](https://openreview.net/forum?id=HygjqjR9Km) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 216 | 6.67 | [Probgan: Towards Probabilistic Gan With Theoretical Guarantees](https://openreview.net/forum?id=H1l7bnR5Ym) | 6, 5, 9 | 1.70 | Accept (Poster) |\n| 217 | 6.67 | [Three Mechanisms Of Weight Decay Regularization](https://openreview.net/forum?id=B1lz-3Rct7) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 218 | 6.67 | [Hierarchical Rl Using An Ensemble Of Proprioceptive Periodic Policies](https://openreview.net/forum?id=SJz1x20cFQ) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 219 | 6.67 | [Detecting Adversarial Examples Via Neural Fingerprinting](https://openreview.net/forum?id=SJekyhCctQ) | 5, 9, 6 | 1.70 | Reject |\n| 220 | 6.67 | [Diversity-sensitive Conditional Generative Adversarial Networks](https://openreview.net/forum?id=rJliMh09F7) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 221 | 6.67 | [Optimal Completion Distillation For Sequence Learning](https://openreview.net/forum?id=rkMW1hRqKX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 222 | 6.67 | [Flowqa: Grasping Flow In History For Conversational Machine Comprehension](https://openreview.net/forum?id=ByftGnR9KX) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 223 | 6.67 | [Towards The First Adversarially Robust Neural Network Model On Mnist](https://openreview.net/forum?id=S1EHOsC9tX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 224 | 6.67 | [Sample Efficient Adaptive Text-to-speech](https://openreview.net/forum?id=rkzjUoAcFX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 225 | 6.67 | [Latent Convolutional Models](https://openreview.net/forum?id=HJGciiR5Y7) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 226 | 6.67 | [Minimal Images In Deep Neural Networks: Fragile Object Recognition In Natural Images](https://openreview.net/forum?id=S1xNb2A9YX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 227 | 6.67 | [Universal Stagewise Learning For Non-convex Problems With Convergence On Averaged Solutions](https://openreview.net/forum?id=Syx5V2CcFm) | 8, 6, 6 | 0.94 | Accept (Poster) |\n| 228 | 6.67 | [Learning Multimodal Graph-to-graph Translation For Molecule Optimization](https://openreview.net/forum?id=B1xJAsA5F7) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 229 | 6.67 | [Autoloss: Learning Discrete Schedule For Alternate Optimization](https://openreview.net/forum?id=BJgK6iA5KX) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 230 | 6.67 | [Efficient Lifelong Learning With A-gem](https://openreview.net/forum?id=Hkf2_sC5FX) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 231 | 6.67 | [Spherical Cnns On Unstructured Grids](https://openreview.net/forum?id=Bkl-43C9FQ) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 232 | 6.67 | [Differentiable Perturb-and-parse: Semi-supervised Parsing With A Structured Variational Autoencoder](https://openreview.net/forum?id=BJlgNh0qKQ) | 8, 7, 5 | 1.25 | Accept (Poster) |\n| 233 | 6.67 | [Practical Lossless Compression With Latent Variables Using Bits Back Coding](https://openreview.net/forum?id=ryE98iR5tm) | 6, 6, 8 | 0.94 | Accept (Poster) |\n| 234 | 6.67 | [Analysis Of Quantized Models](https://openreview.net/forum?id=ryM_IoAqYX) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 235 | 6.67 | [Detecting Memorization In Relu Networks](https://openreview.net/forum?id=HJeB0sC9Fm) | 5, 6, 9 | 1.70 | Reject |\n| 236 | 6.67 | [Snas: Stochastic Neural Architecture Search](https://openreview.net/forum?id=rylqooRqK7) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 237 | 6.67 | [Pate-gan: Generating Synthetic Data With Differential Privacy Guarantees](https://openreview.net/forum?id=S1zk9iRqF7) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 238 | 6.67 | [Principled Deep Neural Network Training Through Linear Programming](https://openreview.net/forum?id=HkMwHsCctm) | 6, 6, 8 | 0.94 | Reject |\n| 239 | 6.67 | [Cot: Cooperative Training For Generative Modeling Of Discrete Data](https://openreview.net/forum?id=SkxxIs0qY7) | 7, 7, 6 | 0.47 | Reject |\n| 240 | 6.67 | [On The Turing Completeness Of Modern Neural Network Architectures](https://openreview.net/forum?id=HyGBdo0qFm) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 241 | 6.67 | [Layoutgan: Generating Graphic Layouts With Wireframe Discriminators](https://openreview.net/forum?id=HJxB5sRcFQ) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 242 | 6.67 | [Learning Factorized Multimodal Representations](https://openreview.net/forum?id=rygqqsA9KX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 243 | 6.67 | [Phase-aware Speech Enhancement With Deep Complex U-net](https://openreview.net/forum?id=SkeRTsAcYm) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 244 | 6.67 | [Go Gradient For Expectation-based Objectives](https://openreview.net/forum?id=ryf6Fs09YX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 245 | 6.67 | [Analyzing Inverse Problems With Invertible Neural Networks](https://openreview.net/forum?id=rJed6j0cKX) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 246 | 6.67 | [Deep Reinforcement Learning With Relational Inductive Biases](https://openreview.net/forum?id=HkxaFoC9KQ) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 247 | 6.67 | [Janossy Pooling: Learning Deep Permutation-invariant Functions For Variable-size Inputs](https://openreview.net/forum?id=BJluy2RcFm) | 7, 5, 8 | 1.25 | Accept (Poster) |\n| 248 | 6.67 | [Improving Generalization And Stability Of Generative Adversarial Networks](https://openreview.net/forum?id=ByxPYjC5KQ) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 249 | 6.67 | [Preconditioner On Matrix Lie Group For Sgd](https://openreview.net/forum?id=Bye5SiAqKX) | 8, 5, 7 | 1.25 | Accept (Poster) |\n| 250 | 6.67 | [Deep Anomaly Detection With Outlier Exposure](https://openreview.net/forum?id=HyxCxhRcY7) | 6, 6, 8 | 0.94 | Accept (Poster) |\n| 251 | 6.67 | [Attention, Learn To Solve Routing Problems!](https://openreview.net/forum?id=ByxBFsRqYm) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 252 | 6.67 | [Learning What And Where To Attend](https://openreview.net/forum?id=BJgLg3R9KQ) | 6, 6, 8 | 0.94 | Accept (Poster) |\n| 253 | 6.67 | [Query-efficient Hard-label Black-box Attack: An Optimization-based Approach](https://openreview.net/forum?id=rJlk6iRqKX) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 254 | 6.67 | [Recall Traces: Backtracking Models For Efficient Reinforcement Learning](https://openreview.net/forum?id=HygsfnR9Ym) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 255 | 6.67 | [Learning To Infer And Execute 3d Shape Programs](https://openreview.net/forum?id=rylNH20qFQ) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 256 | 6.67 | [Dom-q-net: Grounded Rl On Structured Language](https://openreview.net/forum?id=HJgd1nAqFX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 257 | 6.67 | [Toward Understanding The Impact Of Staleness In Distributed Machine Learning](https://openreview.net/forum?id=BylQV305YQ) | 4, 9, 7 | 2.05 | Accept (Poster) |\n| 258 | 6.67 | [Graph Hypernetworks For Neural Architecture Search](https://openreview.net/forum?id=rkgW0oA9FX) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 259 | 6.67 | [A Generative Model For Electron Paths](https://openreview.net/forum?id=r1x4BnCqKX) | 8, 4, 8 | 1.89 | Accept (Poster) |\n| 260 | 6.67 | [Bayesian Prediction Of Future Street Scenes Using Synthetic Likelihoods](https://openreview.net/forum?id=rkgK3oC5Fm) | 6, 8, 6 | 0.94 | Accept (Poster) |\n| 261 | 6.67 | [Disjoint Mapping Network For Cross-modal Matching Of Voices And Faces](https://openreview.net/forum?id=B1exrnCcF7) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 262 | 6.67 | [Complement Objective Training](https://openreview.net/forum?id=HyM7AiA5YX) | 5, 8, 7 | 1.25 | Accept (Poster) |\n| 263 | 6.67 | [Value Propagation Networks](https://openreview.net/forum?id=SJG6G2RqtX) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 264 | 6.67 | [Trellis Networks For Sequence Modeling](https://openreview.net/forum?id=HyeVtoRqtQ) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 265 | 6.67 | [Non-vacuous Generalization Bounds At The Imagenet Scale: A Pac-bayesian Compression Approach](https://openreview.net/forum?id=BJgqqsAct7) | 6, 6, 8 | 0.94 | Accept (Poster) |\n| 266 | 6.67 | [Contingency-aware Exploration In Reinforcement Learning](https://openreview.net/forum?id=HyxGB2AcY7) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 267 | 6.67 | [Context-adaptive Entropy Model For End-to-end Optimized Image Compression](https://openreview.net/forum?id=HyxKIiAqYQ) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 268 | 6.67 | [Learning Finite State Representations Of Recurrent Policy Networks](https://openreview.net/forum?id=S1gOpsCctm) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 269 | 6.67 | [Do Deep Generative Models Know What They Don't Know?](https://openreview.net/forum?id=H1xwNhCcYm) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 270 | 6.67 | [Learning Two-layer Neural Networks With Symmetric Inputs](https://openreview.net/forum?id=H1xipsA5K7) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 271 | 6.67 | [Minimal Random Code Learning: Getting Bits Back From Compressed Model Parameters](https://openreview.net/forum?id=r1f0YiCctm) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 272 | 6.67 | [Noodl: Provable Online Dictionary Learning And Sparse Coding](https://openreview.net/forum?id=HJeu43ActQ) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 273 | 6.67 | [Approximating Cnns With Bag-of-local-features Models Works Surprisingly Well On Imagenet](https://openreview.net/forum?id=SkfMWhAqYQ) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 274 | 6.67 | [Understanding Straight-through Estimator In Training Activation Quantized Neural Nets](https://openreview.net/forum?id=Skh4jRcKQ) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 275 | 6.67 | [Antisymmetricrnn: A Dynamical System View On Recurrent Neural Networks](https://openreview.net/forum?id=ryxepo0cFX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 276 | 6.67 | [The Limitations Of Adversarial Training And The Blind-spot Attack](https://openreview.net/forum?id=HylTBhA5tQ) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 277 | 6.67 | [A Rotation-equivariant Convolutional Neural Network Model Of Primary Visual Cortex](https://openreview.net/forum?id=H1fU8iAqKX) | 7, 5, 8 | 1.25 | Accept (Poster) |\n| 278 | 6.67 | [Generalized Tensor Models For Recurrent Neural Networks](https://openreview.net/forum?id=r1gNni0qtm) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 279 | 6.67 | [Adversarial Attacks On Graph Neural Networks Via Meta Learning](https://openreview.net/forum?id=Bylnx209YX) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 280 | 6.67 | [Training For Faster Adversarial Robustness Verification Via Inducing Relu Stability](https://openreview.net/forum?id=BJfIVjAcKm) | 8, 7, 5 | 1.25 | Accept (Poster) |\n| 281 | 6.67 | [Adv-bnn: Improved Adversarial Defense Through Robust Bayesian Neural Network](https://openreview.net/forum?id=rk4Qso0cKm) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 282 | 6.67 | [Initialized Equilibrium Propagation For Backprop-free Training](https://openreview.net/forum?id=B1GMDsR5tm) | 5, 8, 7 | 1.25 | Accept (Poster) |\n| 283 | 6.67 | [Learning To Design Rna](https://openreview.net/forum?id=ByfyHh05tQ) | 6, 6, 8 | 0.94 | Accept (Poster) |\n| 284 | 6.67 | [Adef: An Iterative Algorithm To Construct Adversarial Deformations](https://openreview.net/forum?id=Hk4dFjR5K7) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 285 | 6.67 | [Stable Opponent Shaping In Differentiable Games](https://openreview.net/forum?id=SyGjjsC5tQ) | 8, 6, 6 | 0.94 | Accept (Poster) |\n| 286 | 6.67 | [Spigan: Privileged Adversarial Learning From Simulation](https://openreview.net/forum?id=rkxoNnC5FQ) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 287 | 6.67 | [Metropolis-hastings View On Variational Inference And Adversarial Training](https://openreview.net/forum?id=Hkg313AcFX) | 5, 6, 9 | 1.70 | Reject |\n| 288 | 6.67 | [Beyond Pixel Norm-balls: Parametric Adversaries Using An Analytically Differentiable Renderer](https://openreview.net/forum?id=SJl2niR9KQ) | 7, 7, 6 | 0.47 | Accept (Poster) |\n| 289 | 6.67 | [Adaptive Posterior Learning: Few-shot Learning With A Surprise-based Memory Module](https://openreview.net/forum?id=ByeSdsC9Km) | 6, 7, 7 | 0.47 | Accept (Poster) |\n| 290 | 6.67 | [Glue: A Multi-task Benchmark And Analysis Platform For Natural Language Understanding](https://openreview.net/forum?id=rJ4km2R5t7) | 7, 5, 8 | 1.25 | Accept (Poster) |\n| 291 | 6.67 | [Looking For Elmo's Friends: Sentence-level Pretraining Beyond Language Modeling](https://openreview.net/forum?id=Bkl87h09FX) | 5, 7, 8 | 1.25 | Reject |\n| 292 | 6.67 | [Misgan: Learning From Incomplete Data With Generative Adversarial Networks](https://openreview.net/forum?id=S1lDV3RcKm) | 7, 6, 7 | 0.47 | Accept (Poster) |\n| 293 | 6.50 | [Gradient Descent Provably Optimizes Over-parameterized Neural Networks](https://openreview.net/forum?id=S1eK3i09YQ) | 3, 8, 8, 7 | 2.06 | Accept (Poster) |\n| 294 | 6.50 | [Relational Forward Models For Multi-agent Learning](https://openreview.net/forum?id=rJlEojAqFm) | 7, 6, 7, 6 | 0.50 | Accept (Poster) |\n| 295 | 6.50 | [Dynamic Channel Pruning: Feature Boosting And Suppression](https://openreview.net/forum?id=BJxh2j0qYm) | 7, 6, 7, 6 | 0.50 | Accept (Poster) |\n| 296 | 6.50 | [Learning Protein Structure With A Differentiable Simulator](https://openreview.net/forum?id=Byg3y3C9Km) | 6, 7, 7, 6 | 0.50 | Accept (Oral) |\n| 297 | 6.50 | [Preferences Implicit In The State Of The World](https://openreview.net/forum?id=rkevMnRqYQ) | 6, 7, 6, 7 | 0.50 | Accept (Poster) |\n| 298 | 6.50 | [Peernets: Exploiting Peer Wisdom Against Adversarial Attacks](https://openreview.net/forum?id=Sk4jFoA9K7) | 7, 6 | 0.50 | Accept (Poster) |\n| 299 | 6.33 | [Multilingual Neural Machine Translation With Soft Decoupled Encoding](https://openreview.net/forum?id=Skeke3C5Fm) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 300 | 6.33 | [Analysing Mathematical Reasoning Abilities Of Neural Models](https://openreview.net/forum?id=H1gR5iR5FX) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 301 | 6.33 | [Minimum Divergence Vs. Maximum Margin: An Empirical Comparison On Seq2seq Models](https://openreview.net/forum?id=H1xD9sR5Fm) | 5, 7, 7 | 0.94 | Accept (Poster) |\n| 302 | 6.33 | [Self-tuning Networks: Bilevel Optimization Of Hyperparameters Using Structured Best-response Functions](https://openreview.net/forum?id=r1eEG20qKQ) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 303 | 6.33 | [Learning Disentangled Representations With Reference-based Variational Autoencoders](https://openreview.net/forum?id=rkxraoRcF7) | 7, 6, 6 | 0.47 | Reject |\n| 304 | 6.33 | [Remember And Forget For Experience Replay](https://openreview.net/forum?id=Bye9LiR9YX) | 7, 6, 6 | 0.47 | Reject |\n| 305 | 6.33 | [Dpsnet: End-to-end Deep Plane Sweep Stereo](https://openreview.net/forum?id=ryeYHi0ctQ) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 306 | 6.33 | [On Tighter Generalization Bounds For Deep Neural Networks: Cnns, Resnets, And Beyond](https://openreview.net/forum?id=SJzwvoCqF7) | 5, 7, 7 | 0.94 | Reject |\n| 307 | 6.33 | [Measuring Compositionality In Representation Learning](https://openreview.net/forum?id=HJz05o0qK7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 308 | 6.33 | [Reward Constrained Policy Optimization](https://openreview.net/forum?id=SkfrvsA9FX) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 309 | 6.33 | [Regularized Learning For Domain Adaptation Under Label Shifts](https://openreview.net/forum?id=rJl0r3R9KX) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 310 | 6.33 | [A Differentiable Self-disambiguated Sense Embedding Model Via Scaled Gumbel Softmax](https://openreview.net/forum?id=Hyls7h05FQ) | 7, 6, 6 | 0.47 | Reject |\n| 311 | 6.33 | [Preventing Posterior Collapse With Delta-vaes](https://openreview.net/forum?id=BJe0Gn0cY7) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 312 | 6.33 | [Efficient Augmentation Via Data Subsampling](https://openreview.net/forum?id=Byxpfh0cFm) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 313 | 6.33 | [Double Viterbi: Weight Encoding For High Compression Ratio And Fast On-chip Reconstruction For Deep Neural Network](https://openreview.net/forum?id=HkfYOoCcYX) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 314 | 6.33 | [Rethinking The Value Of Network Pruning](https://openreview.net/forum?id=rJlnB3C5Ym) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 315 | 6.33 | [Aligning Artificial Neural Networks To The Brain Yields Shallow Recurrent Architectures](https://openreview.net/forum?id=BJeY6sR9KX) | 5, 7, 7 | 0.94 | Reject |\n| 316 | 6.33 | [Equi-normalization Of Neural Networks](https://openreview.net/forum?id=r1gEqiC9FX) | 7, 7, 5 | 0.94 | Accept (Poster) |\n| 317 | 6.33 | [Multi-domain Adversarial Learning](https://openreview.net/forum?id=Sklv5iRqYX) | 5, 8, 6 | 1.25 | Accept (Poster) |\n| 318 | 6.33 | [Information Theoretic Lower Bounds On Negative Log Likelihood](https://openreview.net/forum?id=rkemqsC9Fm) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 319 | 6.33 | [Dialogwae: Multimodal Response Generation With Conditional Wasserstein Auto-encoder](https://openreview.net/forum?id=BkgBvsC9FQ) | 7, 7, 5 | 0.94 | Accept (Poster) |\n| 320 | 6.33 | [Monge-amp\\`ere Flow For Generative Modeling](https://openreview.net/forum?id=rkeUrjCcYQ) | 7, 6, 6 | 0.47 | Reject |\n| 321 | 6.33 | [Nlprolog: Reasoning With Weak Unification For Natural Language Question Answering](https://openreview.net/forum?id=ByfXe2C5tm) | 7, 5, 7 | 0.94 | Reject |\n| 322 | 6.33 | [Attentive Neural Processes](https://openreview.net/forum?id=SkE6PjC9KX) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 323 | 6.33 | [Scalable Unbalanced Optimal Transport Using Generative Adversarial Networks](https://openreview.net/forum?id=HyexAiA5Fm) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 324 | 6.33 | [Structured Neural Summarization](https://openreview.net/forum?id=H1ersoRqtm) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 325 | 6.33 | [Laplacian Networks: Bounding Indicator Function Smoothness For Neural Networks Robustness](https://openreview.net/forum?id=H1e8wsCqYX) | 9, 5, 5 | 1.89 | Reject |\n| 326 | 6.33 | [Accumulation Bit-width Scaling For Ultra-low Precision Training Of Deep Networks](https://openreview.net/forum?id=BklMjsRqY7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 327 | 6.33 | [Direct Optimization Through For Discrete Variational Auto-encoder](https://openreview.net/forum?id=S1ey2sRcYQ) | 7, 7, 5 | 0.94 | Reject |\n| 328 | 6.33 | [Fluctuation-dissipation Relations For Stochastic Gradient Descent](https://openreview.net/forum?id=SkNksoRctQ) | 8, 5, 6 | 1.25 | Accept (Poster) |\n| 329 | 6.33 | [Rnns Implicitly Implement Tensor-product Representations](https://openreview.net/forum?id=BJx0sjC5FX) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 330 | 6.33 | [From Hard To Soft: Understanding Deep Network Nonlinearities Via Vector Quantization And Statistical Inference](https://openreview.net/forum?id=Syxt2jC5FX) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 331 | 6.33 | [Von Mises-fisher Loss For Training Sequence To Sequence Models With Continuous Outputs](https://openreview.net/forum?id=rJlDnoA5Y7) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 332 | 6.33 | [Proxylessnas: Direct Neural Architecture Search On Target Task And Hardware](https://openreview.net/forum?id=HylVB3AqYm) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 333 | 6.33 | [Discriminator Rejection Sampling](https://openreview.net/forum?id=S1GkToR5tm) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 334 | 6.33 | [Visceral Machines: Risk-aversion In Reinforcement Learning With Intrinsic Physiological Rewards](https://openreview.net/forum?id=SyNvti09KQ) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 335 | 6.33 | [Fixup Initialization: Residual Learning Without Normalization](https://openreview.net/forum?id=H1gsz30cKX) | 7, 5, 7 | 0.94 | Accept (Poster) |\n| 336 | 6.33 | [Algorithmic Framework For Model-based Deep Reinforcement Learning With Theoretical Guarantees](https://openreview.net/forum?id=BJe1E2R5KX) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 337 | 6.33 | [Understanding Composition Of Word Embeddings Via Tensor Decomposition](https://openreview.net/forum?id=H1eqjiCctX) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 338 | 6.33 | [Learning To Simulate](https://openreview.net/forum?id=HJgkx2Aqt7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 339 | 6.33 | [Temporal Gaussian Mixture Layer For Videos](https://openreview.net/forum?id=Hyfg5o0qtm) | 6, 6, 7 | 0.47 | Reject |\n| 340 | 6.33 | [Dher: Hindsight Experience Replay For Dynamic Goals](https://openreview.net/forum?id=Byf5-30qFX) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 341 | 6.33 | [L2-nonexpansive Neural Networks](https://openreview.net/forum?id=ByxGSsR9FQ) | 8, 6, 5 | 1.25 | Accept (Poster) |\n| 342 | 6.33 | [Generating Liquid Simulations With Deformation-aware Neural Networks](https://openreview.net/forum?id=HyeGBj09Fm) | 7, 7, 5 | 0.94 | Accept (Poster) |\n| 343 | 6.33 | [Camou: Learning Physical Vehicle Camouflages To Adversarially Attack Detectors In The Wild](https://openreview.net/forum?id=SJgEl3A5tm) | 4, 8, 7 | 1.70 | Accept (Poster) |\n| 344 | 6.33 | [Timbretron: A Wavenet(cyclegan(cqt(audio))) Pipeline For Musical Timbre Transfer](https://openreview.net/forum?id=S1lvm305YQ) | 4, 7, 8 | 1.70 | Accept (Poster) |\n| 345 | 6.33 | [Synthetic Datasets For Neural Program Synthesis](https://openreview.net/forum?id=ryeOSnAqYm) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 346 | 6.33 | [Delta: Deep Learning Transfer Using Feature Map With Attention For Convolutional Networks](https://openreview.net/forum?id=rkgbwsAcYm) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 347 | 6.33 | [Neural Speed Reading With Structural-jump-lstm](https://openreview.net/forum?id=B1xf9jAqFQ) | 7, 5, 7 | 0.94 | Accept (Poster) |\n| 348 | 6.33 | [Policy Generalization In Capacity-limited Reinforcement Learning](https://openreview.net/forum?id=ByxAOoR5K7) | 7, 7, 5 | 0.94 | Reject |\n| 349 | 6.33 | [Large Scale Graph Learning From Smooth Signals](https://openreview.net/forum?id=ryGkSo0qYm) | 7, 5, 7 | 0.94 | Accept (Poster) |\n| 350 | 6.33 | [Post Selection Inference With Incomplete Maximum Mean Discrepancy Estimator](https://openreview.net/forum?id=BkG5SjR5YQ) | 6, 5, 8 | 1.25 | Accept (Poster) |\n| 351 | 6.33 | [Stable Recurrent Models](https://openreview.net/forum?id=Hygxb2CqKm) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 352 | 6.33 | [On The Relation Between The Sharpest Directions Of Dnn Loss And The Sgd Step Length](https://openreview.net/forum?id=SkgEaj05t7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 353 | 6.33 | [Learning To Represent Edits](https://openreview.net/forum?id=BJl6AjC5F7) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 354 | 6.33 | [On Self Modulation For Generative Adversarial Networks](https://openreview.net/forum?id=Hkl5aoR5tm) | 7, 5, 7 | 0.94 | Accept (Poster) |\n| 355 | 6.33 | [Sgd Converges To Global Minimum In Deep Learning Via Star-convex Path](https://openreview.net/forum?id=BylIciRcYQ) | 6, 5, 8 | 1.25 | Accept (Poster) |\n| 356 | 6.33 | [Neural Graph Evolution: Towards Efficient Automatic Robot Design](https://openreview.net/forum?id=BkgWHnR5tm) | 5, 8, 6 | 1.25 | Accept (Poster) |\n| 357 | 6.33 | [The Relativistic Discriminator: A Key Element Missing From Standard Gan](https://openreview.net/forum?id=S1erHoR5t7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 358 | 6.33 | [Augmented Cyclic Adversarial Learning For Low Resource Domain Adaptation](https://openreview.net/forum?id=B1G9doA9F7) | 8, 6, 5 | 1.25 | Accept (Poster) |\n| 359 | 6.33 | [Seq2slate: Re-ranking And Slate Optimization With Rnns](https://openreview.net/forum?id=HkgHk3RctX) | 6, 6, 7 | 0.47 | Reject |\n| 360 | 6.33 | [A Novel Variational Family For Hidden Non-linear Markov Models](https://openreview.net/forum?id=SJMO2iCct7) | 5, 8, 6 | 1.25 | Reject |\n| 361 | 6.33 | [Hierarchical Visuomotor Control Of Humanoids](https://openreview.net/forum?id=BJfYvo09Y7) | 5, 8, 6 | 1.25 | Accept (Poster) |\n| 362 | 6.33 | [Single Shot Neural Architecture Search Via Direct Sparse Optimization](https://openreview.net/forum?id=ryxjH3R5KQ) | 6, 6, 7 | 0.47 | Reject |\n| 363 | 6.33 | [Beyond Greedy Ranking: Slate Optimization Via List-cvae](https://openreview.net/forum?id=r1xX42R5Fm) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 364 | 6.33 | [Local Critic Training Of Deep Neural Networks](https://openreview.net/forum?id=B1x-LjAcKX) | 6, 6, 7 | 0.47 | Reject |\n| 365 | 6.33 | [On The Sensitivity Of Adversarial Robustness To Input Data Distributions](https://openreview.net/forum?id=S1xNEhR9KX) | 7, 5, 7 | 0.94 | Accept (Poster) |\n| 366 | 6.33 | [A Rotation And A Translation Suffice: Fooling Cnns With Simple Transformations](https://openreview.net/forum?id=BJfvknCqFQ) | 8, 6, 5 | 1.25 | Reject |\n| 367 | 6.33 | [Verification Of Non-linear Specifications For Neural Networks](https://openreview.net/forum?id=HyeFAsRctQ) | 7, 5, 7 | 0.94 | Accept (Poster) |\n| 368 | 6.33 | [Visual Reasoning By Progressive Module Networks](https://openreview.net/forum?id=B1fpDsAqt7) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 369 | 6.33 | [Hierarchical Interpretations For Neural Network Predictions](https://openreview.net/forum?id=SkEqro0ctQ) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 370 | 6.33 | [Robust Estimation Via Generative Adversarial Networks](https://openreview.net/forum?id=BJgRDjR9tQ) | 7, 5, 7 | 0.94 | Accept (Poster) |\n| 371 | 6.33 | [Large-scale Answerer In Questioner's Mind For Visual Dialog Question Generation](https://openreview.net/forum?id=rkgT3jRct7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 372 | 6.33 | [Stochastic Gradient Descent Learns State Equations With Nonlinear Activations](https://openreview.net/forum?id=rkeMHjR9Ym) | 7, 5, 7 | 0.94 | Reject |\n| 373 | 6.33 | [Selfless Sequential Learning](https://openreview.net/forum?id=Bkxbrn0cYX) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 374 | 6.33 | [Mae: Mutual Posterior-divergence Regularization For Variational Autoencoders](https://openreview.net/forum?id=Hke4l2AcKQ) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 375 | 6.33 | [Information Asymmetry In Kl-regularized Rl](https://openreview.net/forum?id=S1lqMn05Ym) | 7, 5, 7 | 0.94 | Accept (Poster) |\n| 376 | 6.33 | [Poincare Glove: Hyperbolic Word Embeddings](https://openreview.net/forum?id=Ske5r3AqK7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 377 | 6.33 | [From Language To Goals: Inverse Reinforcement Learning For Vision-based Instruction Following](https://openreview.net/forum?id=r1lq1hRqYQ) | 5, 5, 9 | 1.89 | Accept (Poster) |\n| 378 | 6.33 | [Dynamically Unfolding Recurrent Restorer: A Moving Endpoint Control Method For Image Restoration](https://openreview.net/forum?id=SJfZKiC5FX) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 379 | 6.33 | [Soft Q-learning With Mutual-information Regularization](https://openreview.net/forum?id=HyEtjoCqFX) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 380 | 6.33 | [M^3rl: Mind-aware Multi-agent Management Reinforcement Learning](https://openreview.net/forum?id=BkzeUiRcY7) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 381 | 6.33 | [Invariance And Inverse Stability Under Relu](https://openreview.net/forum?id=SyxYEoA5FX) | 6, 6, 7 | 0.47 | Reject |\n| 382 | 6.33 | [Diversity And Depth In Per-example Routing Models](https://openreview.net/forum?id=BkxWJnC9tX) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 383 | 6.33 | [Revealing Interpretable Object Representations From Human Behavior](https://openreview.net/forum?id=ryxSrhC9KX) | 7, 7, 5 | 0.94 | Accept (Poster) |\n| 384 | 6.33 | [Learning Factorized Representations For Open-set Domain Adaptation](https://openreview.net/forum?id=SJe3HiC5KX) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 385 | 6.33 | [Functional Variational Bayesian Neural Networks](https://openreview.net/forum?id=rkxacs0qY7) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 386 | 6.33 | [Emi: Exploration With Mutual Information Maximizing State And Action Embeddings](https://openreview.net/forum?id=Hylyui09tm) | 5, 7, 7 | 0.94 | Reject |\n| 387 | 6.33 | [Modeling The Long Term Future In Model-based Reinforcement Learning](https://openreview.net/forum?id=SkgQBn0cF7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 388 | 6.33 | [Deepobs: A Deep Learning Optimizer Benchmark Suite](https://openreview.net/forum?id=rJg6ssC5Y7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 389 | 6.33 | [Empirical Bounds On Linear Regions Of Deep Rectifier Networks](https://openreview.net/forum?id=B1MAJhR5YX) | 6, 7, 6 | 0.47 | Reject |\n| 390 | 6.33 | [Signsgd With Majority Vote Is Communication Efficient And Fault Tolerant](https://openreview.net/forum?id=BJxhijAcY7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 391 | 6.33 | [Babyai: A Platform To Study The Sample Efficiency Of Grounded Language Learning](https://openreview.net/forum?id=rJeXCo0cYX) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 392 | 6.33 | [Excessive Invariance Causes Adversarial Vulnerability](https://openreview.net/forum?id=BkfbpsAcF7) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 393 | 6.33 | [Overcoming The Disentanglement Vs Reconstruction Trade-off Via Jacobian Supervision](https://openreview.net/forum?id=Hkg4W2AcFm) | 7, 7, 5 | 0.94 | Accept (Poster) |\n| 394 | 6.33 | [Feature-wise Bias Amplification](https://openreview.net/forum?id=S1ecm2C9K7) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 395 | 6.33 | [Why Do Deep Convolutional Networks Generalize So Poorly To Small Image Transformations?](https://openreview.net/forum?id=HJxYwiC5tm) | 7, 7, 5 | 0.94 | Reject |\n| 396 | 6.33 | [Hierarchical Generative Modeling For Controllable Speech Synthesis](https://openreview.net/forum?id=rygkk305YQ) | 8, 6, 5 | 1.25 | Accept (Poster) |\n| 397 | 6.33 | [Multi-step Retriever-reader Interaction For Scalable Open-domain Question Answering](https://openreview.net/forum?id=HkfPSh05K7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 398 | 6.33 | [Improved Gradient Estimators For Stochastic Discrete Variables](https://openreview.net/forum?id=S1lKSjRcY7) | 7, 6, 6 | 0.47 | Reject |\n| 399 | 6.33 | [Characterizing Audio Adversarial Examples Using Temporal Dependency](https://openreview.net/forum?id=r1g4E3C9t7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 400 | 6.33 | [Data-dependent Coresets For Compressing Neural Networks With Applications To Generalization Bounds](https://openreview.net/forum?id=HJfwJ2A5KX) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 401 | 6.33 | [Meta-learning With Latent Embedding Optimization](https://openreview.net/forum?id=BJgklhAcK7) | 6, 5, 8 | 1.25 | Accept (Poster) |\n| 402 | 6.33 | [Probabilistic Planning With Sequential Monte Carlo Methods](https://openreview.net/forum?id=ByetGn0cYX) | 8, 6, 5 | 1.25 | Accept (Poster) |\n| 403 | 6.33 | [Learning What You Can Do Before Doing Anything](https://openreview.net/forum?id=SylPMnR9Ym) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 404 | 6.33 | [Model-predictive Policy Learning With Uncertainty Regularization For Driving In Dense Traffic](https://openreview.net/forum?id=HygQBn0cYm) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 405 | 6.33 | [Spreading Vectors For Similarity Search](https://openreview.net/forum?id=SkGuG2R5tm) | 6, 7, 6 | 0.47 | Accept (Poster) |\n| 406 | 6.33 | [Learning When To Communicate At Scale In Multiagent Cooperative And Competitive Tasks](https://openreview.net/forum?id=rye7knCqK7) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 407 | 6.33 | [Opportunistic Learning: Budgeted Cost-sensitive Learning From Data Streams](https://openreview.net/forum?id=S1eOHo09KX) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 408 | 6.33 | [The Singular Values Of Convolutional Layers](https://openreview.net/forum?id=rJevYoA9Fm) | 8, 4, 7 | 1.70 | Accept (Poster) |\n| 409 | 6.33 | [Exemplar Guided Unsupervised Image-to-image Translation With Semantic Consistency](https://openreview.net/forum?id=S1lTg3RqYQ) | 6, 5, 8 | 1.25 | Accept (Poster) |\n| 410 | 6.33 | [Learning-based Frequency Estimation Algorithms](https://openreview.net/forum?id=r1lohoCqY7) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 411 | 6.33 | [Max-mig: An Information Theoretic Approach For Joint Learning From Crowds](https://openreview.net/forum?id=BJg9DoR9t7) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 412 | 6.33 | [Multiple-attribute Text Rewriting](https://openreview.net/forum?id=H1g2NhC5KQ) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 413 | 6.33 | [Harmonizing Maximum Likelihood With Gans For Multimodal Conditional Generation](https://openreview.net/forum?id=HJxyAjRcFX) | 8, 7, 4 | 1.70 | Accept (Poster) |\n| 414 | 6.33 | [Variational Autoencoder With Arbitrary Conditioning](https://openreview.net/forum?id=SyxtJh0qYm) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 415 | 6.33 | [A New Dog Learns Old Tricks: Rl Finds Classic Optimization Algorithms](https://openreview.net/forum?id=rkluJ2R9KQ) | 6, 6, 7 | 0.47 | Accept (Poster) |\n| 416 | 6.33 | [Generating Multi-agent Trajectories Using Programmatic Weak Supervision](https://openreview.net/forum?id=rkxw-hAcFQ) | 7, 6, 6 | 0.47 | Accept (Poster) |\n| 417 | 6.25 | [Maximal Divergence Sequential Autoencoder For Binary Software Vulnerability Detection](https://openreview.net/forum?id=ByloIiCqYQ) | 6, 7, 6, 6 | 0.43 | Accept (Poster) |\n| 418 | 6.25 | [Neural Tts Stylization With Adversarial And Collaborative Games](https://openreview.net/forum?id=ByzcS3AcYX) | 6, 6, 6, 7 | 0.43 | Accept (Poster) |\n| 419 | 6.25 | [Competitive Experience Replay](https://openreview.net/forum?id=Sklsm20ctX) | 5, 7, 6, 7 | 0.83 | Accept (Poster) |\n| 420 | 6.25 | [Bayesian Policy Optimization For Model Uncertainty](https://openreview.net/forum?id=SJGvns0qK7) | 5, 6, 7, 7 | 0.83 | Accept (Poster) |\n| 421 | 6.25 | [Sinkhorn Autoencoders](https://openreview.net/forum?id=BygNqoR9tm) | 5, 6, 7, 7 | 0.83 | Reject |\n| 422 | 6.25 | [Two-timescale Networks For Nonlinear Value Function Approximation](https://openreview.net/forum?id=rJleN20qK7) | 6, 7, 6, 6 | 0.43 | Accept (Poster) |\n| 423 | 6.25 | [Lyapunov-based Safe Policy Optimization](https://openreview.net/forum?id=Syxgbh05tQ) | 6, 6, 8, 5 | 1.09 | Reject |\n| 424 | 6.25 | [Towards Consistent Performance On Atari Using Expert Demonstrations](https://openreview.net/forum?id=BkfPnoActQ) | 6, 5, 7, 7 | 0.83 | Reject |\n| 425 | 6.00 | [Learning Implicit Generative Models By Teaching Explicit Ones](https://openreview.net/forum?id=Hkg1YiAcK7) | 7, 5, 6 | 0.82 | Reject |\n| 426 | 6.00 | [Emerging Disentanglement In Auto-encoder Based Unsupervised Image Content Transfer](https://openreview.net/forum?id=BylE1205Fm) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 427 | 6.00 | [Projective Subspace Networks For Few-shot Learning](https://openreview.net/forum?id=rkzfuiA9F7) | 6, 6, 6 | 0.00 | Reject |\n| 428 | 6.00 | [Environment Probing Interaction Policies](https://openreview.net/forum?id=ryl8-3AcFX) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 429 | 6.00 | [Stcn: Stochastic Temporal Convolutional Networks](https://openreview.net/forum?id=HkzSQhCcK7) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 430 | 6.00 | [Capsule Graph Neural Network](https://openreview.net/forum?id=Byl8BnRcYm) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 431 | 6.00 | [Top-down Neural Model For Formulae](https://openreview.net/forum?id=Byg5QhR5FQ) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 432 | 6.00 | [Tarmac: Targeted Multi-agent Communication](https://openreview.net/forum?id=H1e572A5tQ) | 6, 6, 6 | 0.00 | Reject |\n| 433 | 6.00 | [Coarse-grain Fine-grain Coattention Network For Multi-evidence Question Answering](https://openreview.net/forum?id=Syl7OsRqY7) | 7, 7, 4 | 1.41 | Accept (Poster) |\n| 434 | 6.00 | [Learning Programmatically Structured Representations With Perceptor Gradients](https://openreview.net/forum?id=SJggZnRcFQ) | 7, 6, 5 | 0.82 | Accept (Poster) |\n| 435 | 6.00 | [Graph Transformer](https://openreview.net/forum?id=HJei-2RcK7) | 6, 6, 6 | 0.00 | Reject |\n| 436 | 6.00 | [Feed-forward Propagation In Probabilistic Neural Networks With Categorical And Max Layers](https://openreview.net/forum?id=SkMuPjRcKQ) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 437 | 6.00 | [Discriminative Active Learning](https://openreview.net/forum?id=rJl-HsR9KX) | 8, 6, 4 | 1.63 | Reject |\n| 438 | 6.00 | [Neural Logic Machines](https://openreview.net/forum?id=B1xY-hRctX) | 6, 7, 5 | 0.82 | Accept (Poster) |\n| 439 | 6.00 | [Improving Sequence-to-sequence Learning Via Optimal Transport](https://openreview.net/forum?id=S1xtAjR5tX) | 6, 7, 5 | 0.82 | Accept (Poster) |\n| 440 | 6.00 | [Backpropamine: Training Self-modifying Neural Networks With Differentiable Neuromodulated Plasticity](https://openreview.net/forum?id=r1lrAiA5Ym) | 4, 9, 5 | 2.16 | Accept (Poster) |\n| 441 | 6.00 | [Learning To Propagate Labels: Transductive Propagation Network For Few-shot Learning](https://openreview.net/forum?id=SyVuRiC5K7) | 5, 6, 7 | 0.82 | Accept (Poster) |\n| 442 | 6.00 | [Neural Mmo: A Massively Multiplayer Game Environment For Intelligent Agents](https://openreview.net/forum?id=S1gWz2CcKX) | 6, 5, 7 | 0.82 | Reject |\n| 443 | 6.00 | [Learnable Embedding Space For Efficient Neural Architecture Compression](https://openreview.net/forum?id=S1xLN3C9YX) | 5, 7, 6 | 0.82 | Accept (Poster) |\n| 444 | 6.00 | [Ib-gan: Disentangled Representation Learning With Information Bottleneck Gan](https://openreview.net/forum?id=ryljV2A5KX) | 7, 7, 4 | 1.41 | Reject |\n| 445 | 6.00 | [Interpolation-prediction Networks For Irregularly Sampled Time Series](https://openreview.net/forum?id=r1efr3C9Ym) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 446 | 6.00 | [Learning Models For Visual 3d Localization With Implicit Mapping](https://openreview.net/forum?id=rkxusjRctQ) | 7, 5, 6 | 0.82 | Reject |\n| 447 | 6.00 | [Transfer Learning For Related Reinforcement Learning Tasks Via Image-to-image Translation](https://openreview.net/forum?id=rkxjnjA5KQ) | 7, 7, 4 | 1.41 | Reject |\n| 448 | 6.00 | [Countering Language Drift Via Grounding](https://openreview.net/forum?id=BkMn9jAcYQ) | 6, 6, 6 | 0.00 | Reject |\n| 449 | 6.00 | [H-detach: Modifying The Lstm Gradient Towards Better Optimization](https://openreview.net/forum?id=ryf7ioRqFX) | 5, 6, 7 | 0.82 | Accept (Poster) |\n| 450 | 6.00 | [Rigorous Agent Evaluation: An Adversarial Approach To Uncover Catastrophic Failures](https://openreview.net/forum?id=B1xhQhRcK7) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 451 | 6.00 | [Multi-class Classification Without Multi-class Labels](https://openreview.net/forum?id=SJzR2iRcK7) | 6, 7, 5 | 0.82 | Accept (Poster) |\n| 452 | 6.00 | [Dadam: A Consensus-based Distributed Adaptive Gradient Method For Online Optimization](https://openreview.net/forum?id=SJeUAj05tQ) | 8, 4, 6 | 1.63 | Reject |\n| 453 | 6.00 | [A Biologically Inspired Visual Working Memory For Deep Networks](https://openreview.net/forum?id=B1fbosCcYm) | 4, 5, 9 | 2.16 | Reject |\n| 454 | 6.00 | [Multi-agent Dual Learning](https://openreview.net/forum?id=HyGhN2A5tm) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 455 | 6.00 | [Dirichlet Variational Autoencoder](https://openreview.net/forum?id=rkgsvoA9K7) | 6, 5, 7 | 0.82 | Reject |\n| 456 | 6.00 | [Graph Convolutional Network With Sequential Attention For Goal-oriented Dialogue Systems](https://openreview.net/forum?id=Skz-3j05tm) | 5, 6, 7 | 0.82 | Reject |\n| 457 | 6.00 | [Unsupervised Neural Multi-document Abstractive Summarization Of Reviews](https://openreview.net/forum?id=rylhToC5YQ) | 5, 4, 9 | 2.16 | Reject |\n| 458 | 6.00 | [Semi-supervised Learning With Multi-domain Sentiment Word Embeddings](https://openreview.net/forum?id=SkenUj0qYm) | 6, 6, 6 | 0.00 | Reject |\n| 459 | 6.00 | [Guiding Policies With Language Via Meta-learning](https://openreview.net/forum?id=HkgSEnA5KQ) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 460 | 6.00 | [Improving Sentence Representations With Multi-view Frameworks](https://openreview.net/forum?id=S1xzyhR9Y7) | 7, 6, 5 | 0.82 | Reject |\n| 461 | 6.00 | [Estimating Information Flow In Dnns](https://openreview.net/forum?id=HkxOoiAcYX) | 7, 7, 4 | 1.41 | Reject |\n| 462 | 6.00 | [Identifying Bias In Ai Using Simulation](https://openreview.net/forum?id=BJf_YjCqYX) | 5, 7, 6 | 0.82 | Reject |\n| 463 | 6.00 | [Graph Wavelet Neural Network](https://openreview.net/forum?id=H1ewdiR5tQ) | 4, 7, 7 | 1.41 | Accept (Poster) |\n| 464 | 6.00 | [Recurrent Kalman Networks: Factorized Inference In High-dimensional Deep Feature Spaces](https://openreview.net/forum?id=rkx1m2C5YQ) | 6, 6, 6 | 0.00 | Reject |\n| 465 | 6.00 | [Learning Procedural Abstractions And Evaluating Discrete Latent Temporal Structure](https://openreview.net/forum?id=ByleB2CcKm) | 6, 5, 7 | 0.82 | Accept (Poster) |\n| 466 | 6.00 | [Adversarial Information Factorization](https://openreview.net/forum?id=BJfRpoA9YX) | 6, 6, 6 | 0.00 | Reject |\n| 467 | 6.00 | [Bnn+: Improved Binary Network Training](https://openreview.net/forum?id=SJfHg2A5tQ) | 8, 6, 4 | 1.63 | Reject |\n| 468 | 6.00 | [An Empirical Study Of Binary Neural Networks' Optimisation](https://openreview.net/forum?id=rJfUCoR5KX) | 8, 6, 4 | 1.63 | Accept (Poster) |\n| 469 | 6.00 | [Graph U-net](https://openreview.net/forum?id=HJePRoAct7) | 7, 4, 7 | 1.41 | Reject |\n| 470 | 6.00 | [Distribution-interpolation Trade Off In Generative Models](https://openreview.net/forum?id=SyMhLo0qKQ) | 6, 7, 5 | 0.82 | Accept (Poster) |\n| 471 | 6.00 | [A Closer Look At Few-shot Classification](https://openreview.net/forum?id=HkxLXnAcFQ) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 472 | 6.00 | [Decoupled Weight Decay Regularization](https://openreview.net/forum?id=Bkg6RiCqY7) | 6, 7, 5 | 0.82 | Accept (Poster) |\n| 473 | 6.00 | [An Adaptive Homeostatic Algorithm For The Unsupervised Learning Of Visual Features](https://openreview.net/forum?id=SyMras0cFQ) | 5, 4, 9 | 2.16 | Reject |\n| 474 | 6.00 | [Efficient Two-step Adversarial Defense For Deep Neural Networks](https://openreview.net/forum?id=BklpOo09tQ) | 5, 6, 7 | 0.82 | Reject |\n| 475 | 6.00 | [Cramer-wold Autoencoder](https://openreview.net/forum?id=rkgwuiA9F7) | 5, 7, 6 | 0.82 | Reject |\n| 476 | 6.00 | [Precision Highway For Ultra Low-precision Quantization](https://openreview.net/forum?id=SJx94o0qYX) | 6, 7, 5 | 0.82 | Reject |\n| 477 | 6.00 | [Graphseq2seq: Graph-sequence-to-sequence For Neural Machine Translation](https://openreview.net/forum?id=B1fA3oActQ) | 6, 6, 6 | 0.00 | Reject |\n| 478 | 6.00 | [Learning Multi-level Hierarchies With Hindsight](https://openreview.net/forum?id=ryzECoAcY7) | 6, 7, 5 | 0.82 | Accept (Poster) |\n| 479 | 6.00 | [The Variational Deficiency Bottleneck](https://openreview.net/forum?id=rygjN3C9F7) | 5, 7, 6 | 0.82 | Reject |\n| 480 | 6.00 | [Universal Successor Features Approximators](https://openreview.net/forum?id=S1VWjiRcKX) | 7, 5, 6 | 0.82 | Accept (Poster) |\n| 481 | 6.00 | [Deep Lagrangian Networks: Using Physics As Model Prior For Deep Learning](https://openreview.net/forum?id=BklHpjCqKm) | 7, 4, 7 | 1.41 | Accept (Poster) |\n| 482 | 6.00 | [Neural Program Repair By Jointly Learning To Localize And Repair](https://openreview.net/forum?id=ByloJ20qtm) | 6, 7, 5 | 0.82 | Accept (Poster) |\n| 483 | 6.00 | [Measuring And Regularizing Networks In Function Space](https://openreview.net/forum?id=SkMwpiR9Y7) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 484 | 6.00 | [Anytime Minibatch: Exploiting Stragglers In Online Distributed Optimization](https://openreview.net/forum?id=rkzDIiA5YQ) | 4, 7, 7 | 1.41 | Accept (Poster) |\n| 485 | 6.00 | [Stochastic Gradient Push For Distributed Deep Learning](https://openreview.net/forum?id=HkgSk2A9Y7) | 6, 6, 6 | 0.00 | Reject |\n| 486 | 6.00 | [A Direct Approach To Robust Deep Learning Using Adversarial Networks](https://openreview.net/forum?id=S1lIMn05F7) | 5, 7, 6 | 0.82 | Accept (Poster) |\n| 487 | 6.00 | [Gamepad: A Learning Environment For Theorem Proving](https://openreview.net/forum?id=r1xwKoR9Y7) | 4, 7, 7 | 1.41 | Accept (Poster) |\n| 488 | 6.00 | [Don’t Judge A Book By Its Cover - On The Dynamics Of Recurrent Neural Networks](https://openreview.net/forum?id=H1z_Z2A5tX) | 5, 7, 6 | 0.82 | Reject |\n| 489 | 6.00 | [Uncovering Surprising Behaviors In Reinforcement Learning Via Worst-case Analysis](https://openreview.net/forum?id=SkgZNnR5tX) | 5, 7, 6 | 0.82 | Reject |\n| 490 | 6.00 | [Language Model Pre-training For Hierarchical Document Representations](https://openreview.net/forum?id=rygnfn0qF7) | 6, 6, 6 | 0.00 | Reject |\n| 491 | 6.00 | [Manifold Mixup: Learning Better Representations By Interpolating Hidden States](https://openreview.net/forum?id=rJlRKjActQ) | 6, 4, 8 | 1.63 | Reject |\n| 492 | 6.00 | [Dimension-free Bounds For Low-precision Training](https://openreview.net/forum?id=ryeX-nC9YQ) | 6, 6, 6 | 0.00 | Reject |\n| 493 | 6.00 | [Kernel Rnn Learning (kernl)](https://openreview.net/forum?id=ryGfnoC5KQ) | 5, 7, 6 | 0.82 | Accept (Poster) |\n| 494 | 6.00 | [Datnet: Dual Adversarial Transfer For Low-resource Named Entity Recognition](https://openreview.net/forum?id=HkGzUjR5tQ) | 6, 6, 6 | 0.00 | Reject |\n| 495 | 6.00 | [Optimistic Mirror Descent In Saddle-point Problems: Going The Extra (gradient) Mile](https://openreview.net/forum?id=Bkg8jjC9KQ) | 7, 6, 5 | 0.82 | Accept (Poster) |\n| 496 | 6.00 | [Deep Convolutional Networks As Shallow Gaussian Processes](https://openreview.net/forum?id=Bklfsi0cKm) | 5, 8, 5 | 1.41 | Accept (Poster) |\n| 497 | 6.00 | [On The Computational Inefficiency Of Large Batch Sizes For Stochastic Gradient Descent](https://openreview.net/forum?id=S1en0sRqKm) | 5, 8, 5 | 1.41 | Reject |\n| 498 | 6.00 | [Wasserstein Barycenter Model Ensembling](https://openreview.net/forum?id=H1g4k309F7) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 499 | 6.00 | [Computing Committor Functions For The Study Of Rare Events Using Deep Learning With Importance Sampling](https://openreview.net/forum?id=H1lPUiRcYQ) | 6, 6, 5, 7 | 0.71 | Reject |\n| 500 | 6.00 | [Scaling Shared Model Governance Via Model Splitting](https://openreview.net/forum?id=H1xEtoRqtQ) | 4, 5, 9 | 2.16 | Reject |\n| 501 | 6.00 | [Generative Feature Matching Networks](https://openreview.net/forum?id=Syfz6sC9tQ) | 6, 6, 6, 6 | 0.00 | Reject |\n| 502 | 6.00 | [Mixed Precision Quantization Of Convnets Via Differentiable Neural Architecture Search](https://openreview.net/forum?id=BJGVX3CqYm) | 5, 6, 7, 6 | 0.71 | Reject |\n| 503 | 6.00 | [Alignment Based Mathching Networks For One-shot Classification And Open-set Recognition](https://openreview.net/forum?id=Skl6k209Ym) | 7, 6, 7, 4 | 1.22 | Reject |\n| 504 | 6.00 | [Unsupervised Adversarial Image Reconstruction](https://openreview.net/forum?id=BJg4Z3RqF7) | 6, 8, 4 | 1.63 | Accept (Poster) |\n| 505 | 6.00 | [Adversarial Reprogramming Of Neural Networks](https://openreview.net/forum?id=Syx_Ss05tm) | 4, 6, 8 | 1.63 | Accept (Poster) |\n| 506 | 6.00 | [Reinforcement Learning With Perturbed Rewards](https://openreview.net/forum?id=BkMWx309FX) | 6, 6, 6 | 0.00 | Reject |\n| 507 | 6.00 | [Variational Bayesian Phylogenetic Inference](https://openreview.net/forum?id=SJVmjjR9FX) | 6, 5, 7 | 0.82 | Accept (Poster) |\n| 508 | 6.00 | [Efficient Multi-objective Neural Architecture Search Via Lamarckian Evolution](https://openreview.net/forum?id=ByME42AqK7) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 509 | 6.00 | [On-policy Trust Region Policy Optimisation With Replay Buffers](https://openreview.net/forum?id=B1MB5oRqtQ) | 7, 6, 5 | 0.82 | Reject |\n| 510 | 6.00 | [A Max-affine Spline Perspective Of Recurrent Neural Networks](https://openreview.net/forum?id=BJej72AqF7) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 511 | 6.00 | [Explicit Information Placement On Latent Variables Using Auxiliary Generative Modelling Task](https://openreview.net/forum?id=H1l-SjA5t7) | 6, 7, 5 | 0.82 | Reject |\n| 512 | 6.00 | [Code2seq: Generating Sequences From Structured Representations Of Code](https://openreview.net/forum?id=H1gKYo09tX) | 6, 7, 5 | 0.82 | Accept (Poster) |\n| 513 | 6.00 | [Overcoming Catastrophic Forgetting For Continual Learning Via Model Adaptation](https://openreview.net/forum?id=ryGvcoA5YX) | 5, 6, 7 | 0.82 | Accept (Poster) |\n| 514 | 6.00 | [Dl2: Training And Querying Neural Networks With Logic](https://openreview.net/forum?id=H1faSn0qY7) | 7, 5, 6 | 0.82 | Reject |\n| 515 | 6.00 | [Cost-sensitive Robustness Against Adversarial Examples](https://openreview.net/forum?id=BygANhA9tQ) | 5, 5, 8 | 1.41 | Accept (Poster) |\n| 516 | 6.00 | [Robust Conditional Generative Adversarial Networks](https://openreview.net/forum?id=Byg0DsCqYQ) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 517 | 6.00 | [Unsupervised Discovery Of Parts, Structure, And Dynamics](https://openreview.net/forum?id=rJe10iC5K7) | 6, 6, 7, 5 | 0.71 | Accept (Poster) |\n| 518 | 6.00 | [Learning To Learn With Conditional Class Dependencies](https://openreview.net/forum?id=BJfOXnActQ) | 6, 8, 4 | 1.63 | Accept (Poster) |\n| 519 | 6.00 | [Aggregated Momentum: Stability Through Passive Damping](https://openreview.net/forum?id=Syxt5oC5YQ) | 7, 6, 5 | 0.82 | Accept (Poster) |\n| 520 | 6.00 | [Discovery Of Natural Language Concepts In Individual Units Of Cnns](https://openreview.net/forum?id=S1EERs09YQ) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 521 | 6.00 | [Generative Predecessor Models For Sample-efficient Imitation Learning](https://openreview.net/forum?id=SkeVsiAcYm) | 6, 5, 7 | 0.82 | Accept (Poster) |\n| 522 | 6.00 | [Image Deformation Meta-network For One-shot Learning](https://openreview.net/forum?id=Sylw7nCqFQ) | 5, 7, 6 | 0.82 | N/A |\n| 523 | 6.00 | [Adversarial Vulnerability Of Neural Networks Increases With Input Dimension](https://openreview.net/forum?id=H1MzKs05F7) | 6, 4, 9, 5 | 1.87 | Reject |\n| 524 | 6.00 | [How To Train Your Maml](https://openreview.net/forum?id=HJGven05Y7) | 5, 6, 7 | 0.82 | Accept (Poster) |\n| 525 | 6.00 | [Learning Kolmogorov Models For Binary Random Variables](https://openreview.net/forum?id=BJfguoAcFm) | 5, 5, 8 | 1.41 | Reject |\n| 526 | 6.00 | [Unsupervised Hyper-alignment For Multilingual Word Embeddings](https://openreview.net/forum?id=HJe62s09tX) | 5, 6, 7 | 0.82 | Accept (Poster) |\n| 527 | 6.00 | [Adversarial Imitation Via Variational Inverse Reinforcement Learning](https://openreview.net/forum?id=HJlmHoR5tQ) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 528 | 6.00 | [Improving The Generalization Of Adversarial Training With Domain Adaptation](https://openreview.net/forum?id=SyfIfnC5Ym) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 529 | 6.00 | [Formal Limitations On The Measurement Of Mutual Information](https://openreview.net/forum?id=BkedwoC5t7) | 8, 6, 4 | 1.63 | Reject |\n| 530 | 6.00 | [Online Hyperparameter Adaptation Via Amortized Proximal Optimization](https://openreview.net/forum?id=rJl6M2C5Y7) | 6, 5, 7 | 0.82 | Reject |\n| 531 | 6.00 | [Machine Translation With Weakly Paired Bilingual Documents](https://openreview.net/forum?id=ryza73R9tQ) | 6, 5, 7 | 0.82 | Reject |\n| 532 | 6.00 | [Greedy Attack And Gumbel Attack: Generating Adversarial Examples For Discrete Data](https://openreview.net/forum?id=ByghKiC5YX) | 3, 6, 8, 7 | 1.87 | Reject |\n| 533 | 6.00 | [Variance Networks: When Expectation Does Not Meet Your Expectations](https://openreview.net/forum?id=B1GAUs0cKQ) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 534 | 6.00 | [Shallow Learning For Deep Networks](https://openreview.net/forum?id=r1Gsk3R9Fm) | 6, 5, 7 | 0.82 | Reject |\n| 535 | 6.00 | [Success At Any Cost: Value Constrained Model-free Continuous Control](https://openreview.net/forum?id=rJlJ-2CqtX) | 7, 5, 6 | 0.82 | Reject |\n| 536 | 6.00 | [Language Modeling Teaches You More Syntax Than Translation Does: Lessons Learned Through Auxiliary Task Analysis](https://openreview.net/forum?id=ryeNPi0qKX) | 6, 5, 7 | 0.82 | Reject |\n| 537 | 6.00 | [Hierarchical Reinforcement Learning Via Advantage-weighted Information Maximization](https://openreview.net/forum?id=Hyl_vjC5KQ) | 5, 6, 7 | 0.82 | Accept (Poster) |\n| 538 | 6.00 | [Mean-field Analysis Of Batch Normalization](https://openreview.net/forum?id=B1eSg3C9Ym) | 7, 6, 5 | 0.82 | Reject |\n| 539 | 6.00 | [Learning From Positive And Unlabeled Data With A Selection Bias](https://openreview.net/forum?id=rJzLciCqKm) | 7, 6, 5 | 0.82 | Accept (Poster) |\n| 540 | 6.00 | [A Kernel Random Matrix-based Approach For Sparse Pca](https://openreview.net/forum?id=rkgBHoCqYX) | 5, 7, 6 | 0.82 | Accept (Poster) |\n| 541 | 6.00 | [Per-tensor Fixed-point Quantization Of The Back-propagation Algorithm](https://openreview.net/forum?id=rkxaNjA9Ym) | 7, 3, 8 | 2.16 | Accept (Poster) |\n| 542 | 6.00 | [Fortified Networks: Improving The Robustness Of Deep Networks By Modeling The Manifold Of Hidden Representations](https://openreview.net/forum?id=SkgVRiC9Km) | 4, 5, 9, 6 | 1.87 | Reject |\n| 543 | 6.00 | [A Comprehensive, Application-oriented Study Of Catastrophic Forgetting In Dnns](https://openreview.net/forum?id=BkloRs0qK7) | 5, 6, 7 | 0.82 | Accept (Poster) |\n| 544 | 6.00 | [Interactive Agent Modeling By Learning To Probe](https://openreview.net/forum?id=SJl98sR5tX) | 6, 6, 6, 6 | 0.00 | Reject |\n| 545 | 6.00 | [Learning Heuristics For Automated Reasoning Through Reinforcement Learning](https://openreview.net/forum?id=HkeyZhC9F7) | 5, 6, 7 | 0.82 | Reject |\n| 546 | 6.00 | [Stochastic Prediction Of Multi-agent Interactions From Partial Observations](https://openreview.net/forum?id=r1xdH3CcKX) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 547 | 6.00 | [Combinatorial Attacks On Binarized Neural Networks](https://openreview.net/forum?id=S1lTEh09FQ) | 5, 6, 7 | 0.82 | Accept (Poster) |\n| 548 | 6.00 | [Invase: Instance-wise Variable Selection Using Neural Networks](https://openreview.net/forum?id=BJg_roAcK7) | 6, 6, 6 | 0.00 | Accept (Poster) |\n| 549 | 5.75 | [On The Margin Theory Of Feedforward Neural Networks](https://openreview.net/forum?id=HJGtFoC5Fm) | 5, 5, 6, 7 | 0.83 | Reject |\n| 550 | 5.75 | [Neural Networks For Modeling Source Code Edits](https://openreview.net/forum?id=Sklr9i09KQ) | 5, 6, 6, 6 | 0.43 | Reject |\n| 551 | 5.75 | [Efficiently Testing Local Optimality And Escaping Saddles For Relu Networks](https://openreview.net/forum?id=HylTXn0qYX) | 3, 6, 6, 8 | 1.79 | Accept (Poster) |\n| 552 | 5.75 | [Automata Guided Skill Composition](https://openreview.net/forum?id=HkfwpiA9KX) | 7, 5, 6, 5 | 0.83 | Reject |\n| 553 | 5.67 | [A More Globally Accurate Dimensionality Reduction Method Using Triplets](https://openreview.net/forum?id=BkgVx3A9Km) | 6, 5, 6 | 0.47 | Reject |\n| 554 | 5.67 | [Eddi: Efficient Dynamic Discovery Of High-value Information With Partial Vae](https://openreview.net/forum?id=HJl0jiRqtX) | 6, 5, 6 | 0.47 | Reject |\n| 555 | 5.67 | [Deep Imitative Models For Flexible Inference, Planning, And Control](https://openreview.net/forum?id=SyehMhC9Y7) | 5, 6, 6 | 0.47 | Reject |\n| 556 | 5.67 | [Set Transformer](https://openreview.net/forum?id=Hkgnii09Ym) | 5, 6, 6 | 0.47 | Reject |\n| 557 | 5.67 | [Transfer Learning Via Unsupervised Task Discovery For Visual Question Answering](https://openreview.net/forum?id=rkelDoCqFX) | 4, 5, 8 | 1.70 | N/A |\n| 558 | 5.67 | [Information Regularized Neural Networks](https://openreview.net/forum?id=BJgvg30ctX) | 6, 6, 5 | 0.47 | Reject |\n| 559 | 5.67 | [An Information-theoretic Metric Of Transferability For Task Transfer Learning](https://openreview.net/forum?id=BkxAUjRqY7) | 5, 6, 6 | 0.47 | Reject |\n| 560 | 5.67 | [Detecting Out-of-distribution Samples Using Low-order Deep Features Statistics](https://openreview.net/forum?id=rkgpCoRctm) | 5, 5, 7 | 0.94 | Reject |\n| 561 | 5.67 | [Laplacian Smoothing Gradient Descent](https://openreview.net/forum?id=By41BjA9YQ) | 6, 6, 5 | 0.47 | Reject |\n| 562 | 5.67 | [Cross-task Knowledge Transfer For Visually-grounded Navigation](https://openreview.net/forum?id=ByGq7hRqKX) | 7, 5, 5 | 0.94 | Reject |\n| 563 | 5.67 | [Super-resolution Via Conditional Implicit Maximum Likelihood Estimation](https://openreview.net/forum?id=HklyMhCqYQ) | 5, 6, 6 | 0.47 | Reject |\n| 564 | 5.67 | [Mode Normalization](https://openreview.net/forum?id=HyN-M2Rctm) | 5, 6, 6 | 0.47 | Accept (Poster) |\n| 565 | 5.67 | [Learning Cross-lingual Sentence Representations Via A Multi-task Dual-encoder Model](https://openreview.net/forum?id=BJgGhiR5KX) | 7, 4, 6 | 1.25 | Reject |\n| 566 | 5.67 | [Ppo-cma: Proximal Policy Optimization With Covariance Matrix Adaptation](https://openreview.net/forum?id=B1VWtsA5tQ) | 4, 9, 4 | 2.36 | Reject |\n| 567 | 5.67 | [A Resizable Mini-batch Gradient Descent Based On A Multi-armed Bandit](https://openreview.net/forum?id=H1lGHsA9KX) | 6, 7, 4 | 1.25 | Reject |\n| 568 | 5.67 | [Understanding Gans Via Generalization Analysis For Disconnected Support](https://openreview.net/forum?id=B1x5KiCcFX) | 6, 5, 6 | 0.47 | Reject |\n| 569 | 5.67 | [Open-ended Content-style Recombination Via Leakage Filtering](https://openreview.net/forum?id=B1eXbn05t7) | 7, 5, 5 | 0.94 | Reject |\n| 570 | 5.67 | [Adversarial Audio Synthesis](https://openreview.net/forum?id=ByMVTsR5KQ) | 5, 6, 6 | 0.47 | Accept (Poster) |\n| 571 | 5.67 | [A Variational Dirichlet Framework For Out-of-distribution Detection](https://openreview.net/forum?id=ByxmXnA9FQ) | 6, 5, 6 | 0.47 | Reject |\n| 572 | 5.67 | [Stochastic Adversarial Video Prediction](https://openreview.net/forum?id=HyEl3o05Fm) | 6, 6, 5 | 0.47 | Reject |\n| 573 | 5.67 | [Transfer Learning For Sequences Via Learning To Collocate](https://openreview.net/forum?id=ByldlhAqYQ) | 6, 5, 6 | 0.47 | Accept (Poster) |\n| 574 | 5.67 | [Talk The Walk: Navigating Grids In New York City Through Grounded Dialogue](https://openreview.net/forum?id=HyxhusA9Fm) | 6, 7, 4 | 1.25 | Reject |\n| 575 | 5.67 | [Random Mesh Projectors For Inverse Problems](https://openreview.net/forum?id=HyGcghRct7) | 6, 7, 4 | 1.25 | Accept (Poster) |\n| 576 | 5.67 | [Infobot: Transfer And Exploration Via The Information Bottleneck](https://openreview.net/forum?id=rJg8yhAqKm) | 7, 7, 3 | 1.89 | Accept (Poster) |\n| 577 | 5.67 | [Trace-back Along Capsules And Its Application On Semantic Segmentation](https://openreview.net/forum?id=H1xpe2C5Km) | 6, 6, 5 | 0.47 | Reject |\n| 578 | 5.67 | [Adversarially Learned Mixture Model](https://openreview.net/forum?id=SJLhxnRqFQ) | 6, 5, 6 | 0.47 | Reject |\n| 579 | 5.67 | [Unsupervised Document Representation Using Partition Word-vectors Averaging](https://openreview.net/forum?id=HyNbtiR9YX) | 6, 7, 4 | 1.25 | Reject |\n| 580 | 5.67 | [Deep Recurrent Gaussian Process With Variational Sparse Spectrum Approximation](https://openreview.net/forum?id=BkgosiRcKm) | 5, 5, 7 | 0.94 | Reject |\n| 581 | 5.67 | [A Frank-wolfe Framework For Efficient And Effective Adversarial Attacks](https://openreview.net/forum?id=Hyewf3AqYX) | 5, 5, 7 | 0.94 | Reject |\n| 582 | 5.67 | [Adaptive Gradient Methods With Dynamic Bound Of Learning Rate](https://openreview.net/forum?id=Bkg3g2R9FX) | 7, 4, 6 | 1.25 | Accept (Poster) |\n| 583 | 5.67 | [Convolutional Crfs For Semantic Segmentation](https://openreview.net/forum?id=H1xEwsR9FX) | 7, 4, 6 | 1.25 | Reject |\n| 584 | 5.67 | [Ppd: Permutation Phase Defense Against Adversarial Examples In Deep Learning](https://openreview.net/forum?id=HkElFj0qYQ) | 6, 7, 4 | 1.25 | Reject |\n| 585 | 5.67 | [Generalizable Adversarial Training Via Spectral Normalization](https://openreview.net/forum?id=Hyx4knR9Ym) | 6, 5, 6 | 0.47 | Accept (Poster) |\n| 586 | 5.67 | [Cbow Is Not All You Need: Combining Cbow With The Compositional Matrix Space Model](https://openreview.net/forum?id=H1MgjoR9tQ) | 6, 5, 6 | 0.47 | Accept (Poster) |\n| 587 | 5.67 | [Soseleto: A Unified Approach To Transfer Learning And Training With Noisy Labels](https://openreview.net/forum?id=Hye-LiR5Y7) | 7, 5, 5 | 0.94 | Reject |\n| 588 | 5.67 | [Revisiting Reweighted Wake-sleep](https://openreview.net/forum?id=BJzuKiC9KX) | 5, 6, 6 | 0.47 | Reject |\n| 589 | 5.67 | [Lipschitz Regularized Deep Neural Networks Generalize](https://openreview.net/forum?id=r1l3NiCqY7) | 4, 6, 7 | 1.25 | Reject |\n| 590 | 5.67 | [Necst: Neural Joint Source-channel Coding](https://openreview.net/forum?id=BJgbzhC5Ym) | 6, 4, 7 | 1.25 | Reject |\n| 591 | 5.67 | [Multi-objective Training Of Generative Adversarial Networks With Multiple Discriminators](https://openreview.net/forum?id=S1MB-3RcF7) | 6, 5, 6 | 0.47 | Reject |\n| 592 | 5.67 | [Small Steps And Giant Leaps: Minimal Newton Solvers For Deep Learning](https://openreview.net/forum?id=Sygx4305KQ) | 7, 7, 3 | 1.89 | Reject |\n| 593 | 5.67 | [Actrce: Augmenting Experience Via Teacher’s Advice](https://openreview.","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fshaohua0116%2FICLR2019-OpenReviewData","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fshaohua0116%2FICLR2019-OpenReviewData","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fshaohua0116%2FICLR2019-OpenReviewData/lists"}