{"id":13678503,"url":"https://github.com/fajieyuan/SIGIR2021_Conure","last_synced_at":"2025-04-29T15:31:34.601Z","repository":{"id":44559250,"uuid":"363842477","full_name":"fajieyuan/SIGIR2021_Conure","owner":"fajieyuan","description":"Pre-training and Lifelong learning for User Embedding  and Recommender System","archived":false,"fork":false,"pushed_at":"2024-03-15T03:42:37.000Z","size":3610,"stargazers_count":44,"open_issues_count":1,"forks_count":5,"subscribers_count":2,"default_branch":"main","last_synced_at":"2024-11-11T21:37:19.988Z","etag":null,"topics":["catastrophic-forgetting","continual-learning","foundation-model","general-purpose","lifelong-learning","lifelong-machine-learning","one-for-all","pre-training","self-supervised-learning","transfer-learning","universal-recommender","universal-representation","user-profile","user-representation"],"latest_commit_sha":null,"homepage":"","language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":null,"status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/fajieyuan.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":null,"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":"2021-05-03T06:54:17.000Z","updated_at":"2024-10-16T11:51:28.000Z","dependencies_parsed_at":"2024-01-14T15:22:16.201Z","dependency_job_id":"4df47eba-b582-413a-9caa-59943c70a10b","html_url":"https://github.com/fajieyuan/SIGIR2021_Conure","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/fajieyuan%2FSIGIR2021_Conure","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fajieyuan%2FSIGIR2021_Conure/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fajieyuan%2FSIGIR2021_Conure/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/fajieyuan%2FSIGIR2021_Conure/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/fajieyuan","download_url":"https://codeload.github.com/fajieyuan/SIGIR2021_Conure/tar.gz/refs/heads/main","host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":251529550,"owners_count":21603975,"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":["catastrophic-forgetting","continual-learning","foundation-model","general-purpose","lifelong-learning","lifelong-machine-learning","one-for-all","pre-training","self-supervised-learning","transfer-learning","universal-recommender","universal-representation","user-profile","user-representation"],"created_at":"2024-08-02T13:00:54.301Z","updated_at":"2025-04-29T15:31:34.127Z","avatar_url":"https://github.com/fajieyuan.png","language":"Python","funding_links":[],"categories":["Python"],"sub_categories":[],"readme":"# SIGIR2021_Conure\n# One Person, One Model, One World: Learning Continual User Representation without Forgetting\nPosts:\nhttps://zhuanlan.zhihu.com/p/437671278\n\n## Pytorch Code is here: https://github.com/yuangh-x/2022-NIPS-Tenrec\n\n\n🤗 **New Resources**: four Large-scale datasets for evaluating foundation / transferable / cross-domain recommendaiton models.\n\n-  **PixelRec**: https://github.com/westlake-repl/PixelRec\n\n-  **MicroLens(DeepMind Talk)**: https://github.com/westlake-repl/MicroLens\n  \n-  **NineRec(TPAMI)**: https://github.com/westlake-repl/NineRec \n\n-  **Tenrec(NeurIPS)**: https://github.com/yuangh-x/2022-NIPS-Tenrec\n\n\n```\n@inproceedings{yuan2021one,\n  title={One person, one model, one world: Learning continual user representation without forgetting},\n  author={Yuan, Fajie and Zhang, Guoxiao and Karatzoglou, Alexandros and Jose, Joemon and Kong, Beibei and Li, Yudong},\n  booktitle={Proceedings of the 44th International ACM SIGIR Conference on Research and Development in Information Retrieval},\n  pages={696--705},\n  year={2021}\n}\n```\n\n### NextItNet pytorch version: https://github.com/syiswell/NextItNet-Pytorch\n### GRec pytorch version: https://github.com/hangjunguo/GRec\n\n---------------------------------------------------\n\nconure_tp_t1.py: Conure is trained on Task1, and once converged it will be pruned.\n\nconure_ret_t1.py: Conure retrains the pruned architecture of Task1\n\nconure_tp_t2.py: Conure is trained on Task2 and once converged it will be pruned.\n\nconure_ret_t2.py: Conure retrains the pruned architecture of Task2\n\n\n\n\n\n\n## Large-scale Recommendation Dataset for pretraining，transfer learning (crosss-domain recommendation) and user representation learning：\n\nDownload the TTL dataset from:  https://drive.google.com/file/d/1imhHUsivh6oMEtEW-RwVc4OsDqn-xOaP/view?usp=sharin\n            or the ML dataset from: https://drive.google.com/file/d/1-_KmnZFaOdH11keLYVcgkf-kW_BaM266/view?usp=sharing\n\nTTL Dataset: 可用于推荐系统预训练，迁移学习，跨域推荐，冷启动推荐，用户表征学习，自监督学习等任务。\n## Running our code: \nPut these dataset on Data/Session\n\nFOllowing these steps:\n\npython conure_tp_t1.py          After convergence (it takes more than 24 hours for training). We suggest 4 iterations. Parameters will be automatioally saved.\n\npython conure_ret_t1.py          You can manually stop this job if the results are satisfied (better than results reported in conure_tp_t1.py). Parameters will be automatioally saved.\n\npython conure_tp_t2.py\n\npython conure_ret_t2.py \n\npython conure_tp_t3.py\n\npython conure_ret_t3.py  \n\npython conure_tp_t4.py\n\npython conure_ret_t4.py  \n\n\n## Environments\n* Tensorflow (version: 1.10.0)\n* python 2.7\n\n## Related work:\n```\n[1]\n@inproceedings{yuan2019simple,\n  title={A simple convolutional generative network for next item recommendation},\n  author={Yuan, Fajie and Karatzoglou, Alexandros and Arapakis, Ioannis and Jose, Joemon M and He, Xiangnan},\n  booktitle={Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining},\n  pages={582--590},\n  year={2019}\n}\n```\n```\n[2]\n@inproceedings{yuan2020parameter,\n  title={Parameter-efficient transfer from sequential behaviors for user modeling and recommendation},\n  author={Yuan, Fajie and He, Xiangnan and Karatzoglou, Alexandros and Zhang, Liguang},\n  booktitle={Proceedings of the 43rd International ACM SIGIR Conference on Research and Development in Information Retrieval},\n  pages={1469--1478},\n  year={2020}\n}\n```\n```\n[3]\n@inproceedings{yuan2020future,\n  title={Future Data Helps Training: Modeling Future Contexts for Session-based Recommendation},\n  author={Yuan, Fajie and He, Xiangnan and Jiang, Haochuan and Guo, Guibing and Xiong, Jian and Xu, Zhezhao and Xiong, Yilin},\n  booktitle={Proceedings of The Web Conference 2020},\n  pages={303--313},\n  year={2020}\n}\n```\n```\n[4]\n@article{sun2020generic,\n  title={A Generic Network Compression Framework for Sequential Recommender Systems},\n  author={Sun, Yang and Yuan, Fajie and Yang, Ming and Wei, Guoao and Zhao, Zhou and Liu, Duo},\n  journal={Proceedings of the Twelfth ACM International Conference on Web Search and Data Mining},\n  year={2020}\n}\n```\n```\n[5]\n@inproceedings{yuan2016lambdafm,\n  title={Lambdafm: learning optimal ranking with factorization machines using lambda surrogates},\n  author={Yuan, Fajie and Guo, Guibing and Jose, Joemon M and Chen, Long and Yu, Haitao and Zhang, Weinan},\n  booktitle={Proceedings of the 25th ACM International on Conference on Information and Knowledge Management},\n  pages={227--236},\n  year={2016}\n}\n```\n```\n[6]\n@article{wang2020stackrec,\n  title={StackRec: Efficient Training of Very Deep Sequential Recommender Models by Layer Stacking},\n  author={Wang, Jiachun and Yuan, Fajie and Chen, Jian and Wu, Qingyao and Li, Chengmin and Yang, Min and Sun, Yang and Zhang, Guoxiao},\n  journal={arXiv preprint arXiv:2012.07598},\n  year={2020}\n}\n```\n\n#  Hiring\nIf you want to work with Fajie https://fajieyuan.github.io/, Please contact him by email yuanfajie@westlake.edu.cn. His lab is now recruiting visiting students, interns, research assistants, posdocs, and research scientists. You can also contact him if you want to  pursue a Phd degree at Westlake University. \nPlease feel free to talk to him (by weichat: wuxiangwangyuan) if you have ideas or papers for collaboration. He is open to various collaborations. \n西湖大学原发杰团队长期招聘：推荐系统和生物信息（尤其蛋白质相关）方向 ，科研助理，博士生，博后，访问学者，研究员系列。\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffajieyuan%2FSIGIR2021_Conure","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Ffajieyuan%2FSIGIR2021_Conure","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Ffajieyuan%2FSIGIR2021_Conure/lists"}