{"id":13576453,"url":"https://github.com/chao1224/Loss-Balanced-Task-Weighting","last_synced_at":"2025-04-05T05:31:53.395Z","repository":{"id":133803585,"uuid":"149305622","full_name":"chao1224/Loss-Balanced-Task-Weighting","owner":"chao1224","description":"Loss-Balanced Task Weighting to Reduce Negative Transfer in Multi-Task Learning, AAAI-SA'19","archived":false,"fork":false,"pushed_at":"2019-09-23T11:07:41.000Z","size":308,"stargazers_count":29,"open_issues_count":0,"forks_count":7,"subscribers_count":2,"default_branch":"master","last_synced_at":"2025-03-21T02:12:04.882Z","etag":null,"topics":["molecule","multi-task-learning","negative-transfer","transfer-learning"],"latest_commit_sha":null,"homepage":"https://ojs.aaai.org//index.php/AAAI/article/view/5125","language":"Python","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/chao1224.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,"publiccode":null,"codemeta":null}},"created_at":"2018-09-18T14:52:33.000Z","updated_at":"2024-05-20T21:39:24.000Z","dependencies_parsed_at":"2024-07-15T06:02:08.771Z","dependency_job_id":null,"html_url":"https://github.com/chao1224/Loss-Balanced-Task-Weighting","commit_stats":null,"previous_names":[],"tags_count":0,"template":false,"template_full_name":null,"repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chao1224%2FLoss-Balanced-Task-Weighting","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chao1224%2FLoss-Balanced-Task-Weighting/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chao1224%2FLoss-Balanced-Task-Weighting/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/chao1224%2FLoss-Balanced-Task-Weighting/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/chao1224","download_url":"https://codeload.github.com/chao1224/Loss-Balanced-Task-Weighting/tar.gz/refs/heads/master","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":247294449,"owners_count":20915335,"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":["molecule","multi-task-learning","negative-transfer","transfer-learning"],"created_at":"2024-08-01T15:01:10.352Z","updated_at":"2025-04-05T05:31:48.384Z","avatar_url":"https://github.com/chao1224.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# Loss-Balanced Task Weighting\n\nThis repository contains the source code for the paper\n\u003e Shengchao Liu, Yingyu Liang, Anthony Gitter. [Loss-Balanced Task Weighting to Reduce Negative Transfer in Multi-Task Learning\n](https://doi.org/10.1609/aaai.v33i01.33019977). AAAI-SA 2019.\n\nThe figshare appendix is public [here](https://doi.org/10.6084/m9.figshare.7732964).\n\nThis repository implements several multi-task learning algorithms, including the Loss-Balanced Task Weighting (LBTW) approach.\nLBTW dynamically sets tasks weights while training a multi-task neural network.\n\n### Setup\n\n+ [Anaconda2-4.3.1](https://repo.continuum.io/archive/Anaconda2-4.3.1-Linux-x86_64.sh)\n+ PyTorch=0.3\n+ scikit-learn=0.19\n\n### Datasets\n\nRandomly split PubChem BioAssay (PCBA) [1] into 5 folds.\nPCBA has 128 binary classification tasks.\n\n### Experiments\n\n+ 128 Single-Task Learning (STL).\n+ 1 Multi-Task Learning (MTL).\n+ 128 Fine-Tuning.\n+ 2 GradNorm [2].\n+ 1 RMTL [3].\n+ 2 Loss-Balanced Task Weighting (LBTW).\n\n### Result\n\n![Precision-Recall AUC results.](/image/pr.png)\n\n### Reference\n\n[1] Wang, Yanli, Stephen H. Bryant, Tiejun Cheng, Jiyao Wang, Asta Gindulyte, Benjamin A. Shoemaker, Paul A. Thiessen, Siqian He, and Jian Zhang. \"Pubchem bioassay: 2017 update.\" Nucleic acids research 45, no. D1: D955-D963, 2016.\n\n[2] Chen, Zhao, Vijay Badrinarayanan, Chen-Yu Lee, and Andrew Rabinovich. \"Gradnorm: Gradient normalization for adaptive loss balancing in deep multitask networks.\" arXiv preprint arXiv:1711.02257, 2017.\n\n[3] Liu, Shengchao. \"Exploration on Deep Drug Discovery: Representation and Learning.\" University of Wisconsin-Madison, Master’s Thesis TR1854, 2018.\n\n### License\n\nThis project is released under the [MIT license](https://github.com/chao1224/Loss-Balanced-Task-Weighting/blob/master/LICENSE).\n\n### Citation\n\n```\n@article{liu2019loss,\n    title={Loss-Balanced Task Weighting to Reduce Negative Transfer in Multi-Task Learning},\n    author={Liu, Shengchao and Liang, Yingyu and Gitter, Anthony},\n    booktitle={Association for the Advancement of Artificial Intelligence (Student Abstract)},\n    year={2019}\n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchao1224%2FLoss-Balanced-Task-Weighting","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fchao1224%2FLoss-Balanced-Task-Weighting","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fchao1224%2FLoss-Balanced-Task-Weighting/lists"}