{"id":22229756,"url":"https://github.com/graph-com/sgd_unlearning","last_synced_at":"2026-03-18T02:35:22.860Z","repository":{"id":259623149,"uuid":"744606663","full_name":"Graph-COM/SGD_unlearning","owner":"Graph-COM","description":null,"archived":false,"fork":false,"pushed_at":"2024-10-26T22:09:09.000Z","size":25343,"stargazers_count":4,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2025-07-17T06:48:12.228Z","etag":null,"topics":[],"latest_commit_sha":null,"homepage":null,"language":"Jupyter Notebook","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/Graph-COM.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":"2024-01-17T16:37:26.000Z","updated_at":"2025-01-13T08:39:24.000Z","dependencies_parsed_at":"2024-10-27T01:26:47.029Z","dependency_job_id":null,"html_url":"https://github.com/Graph-COM/SGD_unlearning","commit_stats":null,"previous_names":["graph-com/sgd_unlearning"],"tags_count":0,"template":false,"template_full_name":null,"purl":"pkg:github/Graph-COM/SGD_unlearning","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Graph-COM%2FSGD_unlearning","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Graph-COM%2FSGD_unlearning/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Graph-COM%2FSGD_unlearning/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Graph-COM%2FSGD_unlearning/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/Graph-COM","download_url":"https://codeload.github.com/Graph-COM/SGD_unlearning/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/Graph-COM%2FSGD_unlearning/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":273123556,"owners_count":25049830,"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","status":"online","status_checked_at":"2025-09-01T02:00:09.058Z","response_time":120,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"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":[],"created_at":"2024-12-03T01:12:14.967Z","updated_at":"2026-03-18T02:35:17.831Z","avatar_url":"https://github.com/Graph-COM.png","language":"Jupyter Notebook","funding_links":[],"categories":[],"sub_categories":[],"readme":"# SGD_Unlearning\n\nThis is the official implementation of **Neurips 2024** paper \n\n[Certified Machine Unlearning via Noisy Stochastic Gradient Descent](https://arxiv.org/abs/2403.17105)\n\nBy Eli Chien, Haoyu Wang, Ziang Chen and Pan Li.\n\nPlease also check our **NeurIPS 2024 Spotlight** paper\n\n[Langevin Unlearning: A New Perspective of Noisy Gradient Descent for Machine Unlearning](https://arxiv.org/abs/2401.10371)\n\n## Environment requirements\n\nThe code is runnable under the following enveironment:\n\n````\nmatplotlib                      3.7.2\nnotebook                        7.0.7\nnumpy                           1.24.4\npandas                          2.0.3\nscikit-learn                    1.3.0\nscipy                           1.10.1\nseaborn                         0.13.0\ntorch                           2.0.0+cu117\ntorchvision                     0.15.1+cu117\ntqdm                            4.65.0\n````\n\n## To implement and re-produce the result in Figure 3.a, run\n\n````\npython main_sgd.py --lam 1e-6 --dataset [MNIST/CIFAR10] --projection 0 --compare_baseline_nonconvergent 1\n````\n\n## To implement and re-produce the result in Figure 3.b, run\n\n````\npython main_sgd.py --lam 1e-6 --dataset [MNIST/CIFAR10] --projection 0 --sequential 1\n````\n\n## To implement and re-produce the result in Figure 3.c.d, run\n\n````\npython main_sgd.py --lam 1e-6 --dataset [MNIST/CIFAR10] --projection 0 --paint_unlearning_sigma 1\n````\n\n## Note\n\n* use --gpu to allocate to a GPU device\n\n* /result/SGD saves the results we run and report in our paper\n\n\n## Citation\n\nIf you find our work useful, please cite us:\n```\n@misc{chien2024certifiedmachineunlearningnoisy,\n      title={Certified Machine Unlearning via Noisy Stochastic Gradient Descent}, \n      author={Eli Chien and Haoyu Wang and Ziang Chen and Pan Li},\n      year={2024},\n      eprint={2403.17105},\n      archivePrefix={arXiv},\n      primaryClass={cs.LG},\n      url={https://arxiv.org/abs/2403.17105}, \n}\n```\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgraph-com%2Fsgd_unlearning","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fgraph-com%2Fsgd_unlearning","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fgraph-com%2Fsgd_unlearning/lists"}