{"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/identifying-and-compensating-for-feature","title":"Identifying and Compensating for Feature Deviation in Imbalanced Deep Learning","arxiv_id":"2001.01385","date":"2020-01-06","proceeding":null,"authors":["Han-Jia Ye","Hong-You Chen","De-Chuan Zhan","Wei-Lun Chao"],"abstract":"Classifiers trained with class-imbalanced data are known to perform poorly on test data of the \"minor\" classes, of which we have insufficient training data. In this paper, we investigate learning a ConvNet classifier under such a scenario. We found that a ConvNet significantly over-fits the minor classes, which is quite opposite to traditional machine learning algorithms that often under-fit minor classes. We conducted a series of analysis and discovered the feature deviation phenomenon -- the learned ConvNet generates deviated features between the training and test data of minor classes -- which explains how over-fitting happens. To compensate for the effect of feature deviation which pushes test data toward low decision value regions, we propose to incorporate class-dependent temperatures (CDT) in training a ConvNet. CDT simulates feature deviation in the training phase, forcing the ConvNet to enlarge the decision values for minor-class data so that it can overcome real feature deviation in the test phase. We validate our approach on benchmark datasets and achieve promising performance. We hope that our insights can inspire new ways of thinking in resolving class-imbalanced deep learning.","url_abs":"https://arxiv.org/abs/2001.01385v4","url_pdf":"https://arxiv.org/pdf/2001.01385v4.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":"identifying-and-compensating-for-feature","repo_url":"https://github.com/zhangyongshun/BagofTricks-LT","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"deep-learning","task_name":"Deep Learning"}],"methods":[{"method_slug":"logistic-regression","method_name":"Logistic Regression"},{"method_slug":"test","method_name":"Test"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2001.01385","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2001.01385"}},"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/zhangyongshun/BagofTricks-LT","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"unverified":7},"by_repo_kind":{"listed":{"samples":8,"ran":1,"repositories":1}},"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":0,"samples":[{"code_sha256_prefix":"fac5364e2f53c6db","entry":"conv3x3","repo":"zhangyongshun/BagofTricks-LT","repo_kind":"listed","path":"lib/backbone/oltr_resnet.py","file_url":"https://github.com/zhangyongshun/BagofTricks-LT/blob/HEAD/lib/backbone/oltr_resnet.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"fac5364e2f53c6db"}},{"code_sha256_prefix":"20991a2f2993c1e7","entry":"accuracy","repo":"zhangyongshun/BagofTricks-LT","repo_kind":"listed","path":"lib/core/evaluate.py","file_url":"https://github.com/zhangyongshun/BagofTricks-LT/blob/HEAD/lib/core/evaluate.py","link_basis":"harvester_set","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":"20991a2f2993c1e7"}},{"code_sha256_prefix":"4826cd3596c12660","entry":"init_weights","repo":"zhangyongshun/BagofTricks-LT","repo_kind":"listed","path":"lib/backbone/oltr_resnet.py","file_url":"https://github.com/zhangyongshun/BagofTricks-LT/blob/HEAD/lib/backbone/oltr_resnet.py","link_basis":"harvester_set","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":"4826cd3596c12660"}},{"code_sha256_prefix":"75000f8601e0929d","entry":"reduce_tensor","repo":"zhangyongshun/BagofTricks-LT","repo_kind":"listed","path":"lib/core/function.py","file_url":"https://github.com/zhangyongshun/BagofTricks-LT/blob/HEAD/lib/core/function.py","link_basis":"harvester_set","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":"75000f8601e0929d"}},{"code_sha256_prefix":"cd77b0f5ea846342","entry":"res10","repo":"zhangyongshun/BagofTricks-LT","repo_kind":"listed","path":"lib/backbone/oltr_resnet.py","file_url":"https://github.com/zhangyongshun/BagofTricks-LT/blob/HEAD/lib/backbone/oltr_resnet.py","link_basis":"harvester_set","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":"cd77b0f5ea846342"}},{"code_sha256_prefix":"4da6f52731511b02","entry":"res32_cifar","repo":"zhangyongshun/BagofTricks-LT","repo_kind":"listed","path":"lib/backbone/resnet_cifar.py","file_url":"https://github.com/zhangyongshun/BagofTricks-LT/blob/HEAD/lib/backbone/resnet_cifar.py","link_basis":"harvester_set","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":"4da6f52731511b02"}},{"code_sha256_prefix":"110330294bf78ea2","entry":"res50","repo":"zhangyongshun/BagofTricks-LT","repo_kind":"listed","path":"lib/backbone/resnet.py","file_url":"https://github.com/zhangyongshun/BagofTricks-LT/blob/HEAD/lib/backbone/resnet.py","link_basis":"harvester_set","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":"110330294bf78ea2"}},{"code_sha256_prefix":"bdb1deaaf5078648","entry":"resnext50","repo":"zhangyongshun/BagofTricks-LT","repo_kind":"listed","path":"lib/backbone/resnext.py","file_url":"https://github.com/zhangyongshun/BagofTricks-LT/blob/HEAD/lib/backbone/resnext.py","link_basis":"harvester_set","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":"bdb1deaaf5078648"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}