{"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/decoupled-training-for-long-tailed","title":"Decoupled Training for Long-Tailed Classification With Stochastic Representations","arxiv_id":"2304.09426","date":"2023-04-19","proceeding":null,"authors":["Giung Nam","Sunguk Jang","Juho Lee"],"abstract":"Decoupling representation learning and classifier learning has been shown to be effective in classification with long-tailed data. There are two main ingredients in constructing a decoupled learning scheme; 1) how to train the feature extractor for representation learning so that it provides generalizable representations and 2) how to re-train the classifier that constructs proper decision boundaries by handling class imbalances in long-tailed data. In this work, we first apply Stochastic Weight Averaging (SWA), an optimization technique for improving the generalization of deep neural networks, to obtain better generalizing feature extractors for long-tailed classification. We then propose a novel classifier re-training algorithm based on stochastic representation obtained from the SWA-Gaussian, a Gaussian perturbed SWA, and a self-distillation strategy that can harness the diverse stochastic representations based on uncertainty estimates to build more robust classifiers. Extensive experiments on CIFAR10/100-LT, ImageNet-LT, and iNaturalist-2018 benchmarks show that our proposed method improves upon previous methods both in terms of prediction accuracy and uncertainty estimation.","url_abs":"https://arxiv.org/abs/2304.09426v1","url_pdf":"https://arxiv.org/pdf/2304.09426v1.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":[],"tasks":[{"task_slug":"classification-1","task_name":"Classification"},{"task_slug":"representation-learning","task_name":"Representation Learning"}],"methods":[{"method_slug":"stochastic-weight-averaging","method_name":"Stochastic Weight Averaging"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2304.09426","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2304.09426"}},"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":"deterministic:regex_extraction","url":"https://github.com/cs-giung/long-tailed-srepr","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR","reach":{"status":"ok","spdx":"BSD-3-Clause"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/visipedia/inat","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/kaidic/LDAM-DRW","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_fixture":1,"ran_draft_wrong":2,"unverified":11},"by_repo_kind":{"found_in_text":{"samples":14,"ran":3,"repositories":2}},"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":"1e369e39c90cada6","entry":"accuracy","repo":"kaidic/LDAM-DRW","repo_kind":"found_in_text","path":"utils.py","file_url":"https://github.com/kaidic/LDAM-DRW/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"1e369e39c90cada6"}},{"code_sha256_prefix":"fac5364e2f53c6db","entry":"conv3x3","repo":"zhmiao/OpenLongTailRecognition-OLTR","repo_kind":"found_in_text","path":"models/ResNetFeature.py","file_url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR/blob/HEAD/models/ResNetFeature.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":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"fac5364e2f53c6db"}},{"code_sha256_prefix":"4dcf06ba28983266","entry":"focal_loss","repo":"kaidic/LDAM-DRW","repo_kind":"found_in_text","path":"losses.py","file_url":"https://github.com/kaidic/LDAM-DRW/blob/HEAD/losses.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4dcf06ba28983266"}},{"code_sha256_prefix":"84954cc67863b073","entry":"create_loss","repo":"zhmiao/OpenLongTailRecognition-OLTR","repo_kind":"found_in_text","path":"loss/DiscCentroidsLoss.py","file_url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR/blob/HEAD/loss/DiscCentroidsLoss.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"84954cc67863b073"}},{"code_sha256_prefix":"a50d9d62804228f2","entry":"create_model","repo":"zhmiao/OpenLongTailRecognition-OLTR","repo_kind":"found_in_text","path":"models/CosNormClassifier.py","file_url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR/blob/HEAD/models/CosNormClassifier.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"a50d9d62804228f2"}},{"code_sha256_prefix":"a2ab1cdb1975f18e","entry":"create_model","repo":"zhmiao/OpenLongTailRecognition-OLTR","repo_kind":"found_in_text","path":"models/DotProductClassifier.py","file_url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR/blob/HEAD/models/DotProductClassifier.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"a2ab1cdb1975f18e"}},{"code_sha256_prefix":"aec4f3ee8eb5c0e4","entry":"create_model","repo":"zhmiao/OpenLongTailRecognition-OLTR","repo_kind":"found_in_text","path":"models/ResNet10Feature.py","file_url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR/blob/HEAD/models/ResNet10Feature.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"aec4f3ee8eb5c0e4"}},{"code_sha256_prefix":"9b5c18aaeaac34a0","entry":"create_model","repo":"zhmiao/OpenLongTailRecognition-OLTR","repo_kind":"found_in_text","path":"models/ResNet152Feature.py","file_url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR/blob/HEAD/models/ResNet152Feature.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"9b5c18aaeaac34a0"}},{"code_sha256_prefix":"4826cd3596c12660","entry":"init_weights","repo":"zhmiao/OpenLongTailRecognition-OLTR","repo_kind":"found_in_text","path":"utils.py","file_url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR/blob/HEAD/utils.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"4826cd3596c12660"}},{"code_sha256_prefix":"e5ce72a1e8d9bc22","entry":"plot_confusion_matrix","repo":"kaidic/LDAM-DRW","repo_kind":"found_in_text","path":"utils.py","file_url":"https://github.com/kaidic/LDAM-DRW/blob/HEAD/utils.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":"e5ce72a1e8d9bc22"}},{"code_sha256_prefix":"11d038f53d77a8d2","entry":"resnet32","repo":"kaidic/LDAM-DRW","repo_kind":"found_in_text","path":"models/resnet_cifar.py","file_url":"https://github.com/kaidic/LDAM-DRW/blob/HEAD/models/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":"11d038f53d77a8d2"}},{"code_sha256_prefix":"ebb474138acbc60c","entry":"shot_acc","repo":"zhmiao/OpenLongTailRecognition-OLTR","repo_kind":"found_in_text","path":"utils.py","file_url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR/blob/HEAD/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"ebb474138acbc60c"}},{"code_sha256_prefix":"07cca9631fe4f26a","entry":"source_import","repo":"zhmiao/OpenLongTailRecognition-OLTR","repo_kind":"found_in_text","path":"utils.py","file_url":"https://github.com/zhmiao/OpenLongTailRecognition-OLTR/blob/HEAD/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"BSD-3-Clause","inline_ok":true,"mcp_get_code":{"code_sha256":"07cca9631fe4f26a"}},{"code_sha256_prefix":"d9855945f564e8f2","entry":"validate","repo":"kaidic/LDAM-DRW","repo_kind":"found_in_text","path":"cifar_train.py","file_url":"https://github.com/kaidic/LDAM-DRW/blob/HEAD/cifar_train.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":"d9855945f564e8f2"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}