{"about":{"site":"https://codewithpapers.app","non_affiliation":"Code with Papers and Syntology are not affiliated with, endorsed by, or sponsored by Papers with Code, Meta, or the pwc-archive mirror.","licence":"CC BY-SA 4.0","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","attribution":"https://codewithpapers.app/attribution","modified":"archive material modified by Syntology; see the attribution page"},"url":"/paper/learning-deep-representations-by-mutual","title":"Learning deep representations by mutual information estimation and maximization","arxiv_id":"1808.06670","date":"2018-08-20","proceeding":"ICLR 2019 5","authors":["R. Devon Hjelm","Alex Fedorov","Samuel Lavoie-Marchildon","Karan Grewal","Phil Bachman","Adam Trischler","Yoshua Bengio"],"abstract":"In this work, we perform unsupervised learning of representations by\nmaximizing mutual information between an input and the output of a deep neural\nnetwork encoder. Importantly, we show that structure matters: incorporating\nknowledge about locality of the input to the objective can greatly influence a\nrepresentation's suitability for downstream tasks. We further control\ncharacteristics of the representation by matching to a prior distribution\nadversarially. Our method, which we call Deep InfoMax (DIM), outperforms a\nnumber of popular unsupervised learning methods and competes with\nfully-supervised learning on several classification tasks. DIM opens new\navenues for unsupervised learning of representations and is an important step\ntowards flexible formulations of representation-learning objectives for\nspecific end-goals.","url_abs":"http://arxiv.org/abs/1808.06670v5","url_pdf":"http://arxiv.org/pdf/1808.06670v5.pdf","source":{"archive":"pwc-archive (Hugging Face), CC BY-SA 4.0","snapshot":"2025-07-28","licence_url":"https://creativecommons.org/licenses/by-sa/4.0/legalcode","row_kind":"abstracts"},"code_links":[{"paper_slug":"learning-deep-representations-by-mutual","repo_url":"https://github.com/rdevon/DIM","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-deep-representations-by-mutual","repo_url":"https://github.com/DuaneNielsen/DeepInfomaxPytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-deep-representations-by-mutual","repo_url":"https://github.com/HolenYHR/Deepinfo_pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"learning-deep-representations-by-mutual","repo_url":"https://github.com/bojone/infomax","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"learning-deep-representations-by-mutual","repo_url":"https://github.com/createamind/DIM_Commented","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"unanswered"}},{"paper_slug":"learning-deep-representations-by-mutual","repo_url":"https://github.com/ifding/simple-Infomax-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"learning-deep-representations-by-mutual","repo_url":"https://github.com/jqhoogland/rgpy","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"learning-deep-representations-by-mutual","repo_url":"https://github.com/schzhu/learning-adversarially-robust-representations","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":null}],"tasks":[{"task_slug":"classification","task_name":"General Classification"},{"task_slug":"mutual-information-estimation","task_name":"Mutual Information Estimation"},{"task_slug":"representation-learning","task_name":"Representation Learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1808.06670","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1808.06670"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/schzhu/learning-adversarially-robust-representations","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/ifding/simple-Infomax-pytorch","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/rdevon/DIM","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/jqhoogland/rgpy","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/HolenYHR/Deepinfo_pytorch","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/DuaneNielsen/DeepInfomaxPytorch","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/bojone/infomax","reach":{"status":"unanswered"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/createamind/DIM_Commented","reach":{"status":"unanswered"}}],"summary":{"ran":4,"ran_fixture":1,"unverified":1},"by_repo_kind":{"listed":{"samples":6,"ran":5,"repositories":3}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":2,"samples":[{"code_sha256_prefix":"fa2d8d0740b5f178","entry":"DeepInfoMaxLoss","repo":"ifding/simple-Infomax-pytorch","repo_kind":"listed","path":"models.py","file_url":"https://github.com/ifding/simple-Infomax-pytorch/blob/HEAD/models.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fa2d8d0740b5f178"}},{"code_sha256_prefix":"34defba22049fee0","entry":"Encoder","repo":"HolenYHR/Deepinfo_pytorch","repo_kind":"listed","path":"model.py","file_url":"https://github.com/HolenYHR/Deepinfo_pytorch/blob/HEAD/model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"34defba22049fee0"}},{"code_sha256_prefix":"9294e9272cea78ad","entry":"LocalDiscriminator","repo":"ifding/simple-Infomax-pytorch","repo_kind":"listed","path":"models.py","file_url":"https://github.com/ifding/simple-Infomax-pytorch/blob/HEAD/models.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9294e9272cea78ad"}},{"code_sha256_prefix":"727245272a3c5183","entry":"PriorDiscriminator","repo":"ifding/simple-Infomax-pytorch","repo_kind":"listed","path":"models.py","file_url":"https://github.com/ifding/simple-Infomax-pytorch/blob/HEAD/models.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"727245272a3c5183"}},{"code_sha256_prefix":"2ca8bbd852e6a9f2","entry":"infonce_loss","repo":"schzhu/learning-adversarially-robust-representations","repo_kind":"listed","path":"robust_representations/functions/dim_losses.py","file_url":"https://github.com/schzhu/learning-adversarially-robust-representations/blob/HEAD/robust_representations/functions/dim_losses.py","link_basis":"first_harvest_node","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"2ca8bbd852e6a9f2"}},{"code_sha256_prefix":"2eade593e94bf4be","entry":"GlobalDiscriminator","repo":"ifding/simple-Infomax-pytorch","repo_kind":"listed","path":"models.py","file_url":"https://github.com/ifding/simple-Infomax-pytorch/blob/HEAD/models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2eade593e94bf4be"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}