{"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/self-knowledge-distillation-a-simple-way-for","title":"Self-Knowledge Distillation with Progressive Refinement of Targets","arxiv_id":"2006.12000","date":"2020-06-22","proceeding":"ICCV 2021 10","authors":["Kyungyul Kim","ByeongMoon Ji","Doyoung Yoon","Sangheum Hwang"],"abstract":"The generalization capability of deep neural networks has been substantially improved by applying a wide spectrum of regularization methods, e.g., restricting function space, injecting randomness during training, augmenting data, etc. In this work, we propose a simple yet effective regularization method named progressive self-knowledge distillation (PS-KD), which progressively distills a model's own knowledge to soften hard targets (i.e., one-hot vectors) during training. Hence, it can be interpreted within a framework of knowledge distillation as a student becomes a teacher itself. Specifically, targets are adjusted adaptively by combining the ground-truth and past predictions from the model itself. We show that PS-KD provides an effect of hard example mining by rescaling gradients according to difficulty in classifying examples. The proposed method is applicable to any supervised learning tasks with hard targets and can be easily combined with existing regularization methods to further enhance the generalization performance. Furthermore, it is confirmed that PS-KD achieves not only better accuracy, but also provides high quality of confidence estimates in terms of calibration as well as ordinal ranking. Extensive experimental results on three different tasks, image classification, object detection, and machine translation, demonstrate that our method consistently improves the performance of the state-of-the-art baselines. The code is available at https://github.com/lgcnsai/PS-KD-Pytorch.","url_abs":"https://arxiv.org/abs/2006.12000v3","url_pdf":"https://arxiv.org/pdf/2006.12000v3.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":"self-knowledge-distillation-a-simple-way-for","repo_url":"https://github.com/lgcnsai/ps-kd-pytorch","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-classification","task_name":"Image Classification"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"},{"task_slug":"machine-translation","task_name":"Machine Translation"},{"task_slug":"multimodal-machine-translation","task_name":"Multimodal Machine Translation"},{"task_slug":"object-detection","task_name":"Object Detection"},{"task_slug":"self-knowledge-distillation","task_name":"Self-Knowledge Distillation"},{"task_slug":"image-classification","task_name":"image-classification"},{"task_slug":"object-detection-1","task_name":"object-detection"}],"methods":[{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-classification-on-cifar-100","task":"Image Classification","dataset":"CIFAR-100","model":"PyramidNet-200 + Shakedrop + Cutmix + PS-KD","rank_in_archive_order":52,"of":211,"metrics":{"Percentage correct":"86.41"},"uses_additional_data":true},{"leaderboard":"/sota/image-classification-on-imagenet","task":"Image Classification","dataset":"ImageNet","model":"PS-KD (ResNet-152 + CutMix)","rank_in_archive_order":768,"of":1060,"metrics":{"Top 1 Accuracy":"79.24%"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-iwslt2015-english","task":"Machine Translation","dataset":"IWSLT2015 English-German","model":"PS-KD","rank_in_archive_order":1,"of":8,"metrics":{"BLEU score":"30.00"},"uses_additional_data":false},{"leaderboard":"/sota/machine-translation-on-iwslt2015-german","task":"Machine Translation","dataset":"IWSLT2015 German-English","model":"PS-KD","rank_in_archive_order":1,"of":15,"metrics":{"BLEU score":"36.20"},"uses_additional_data":false},{"leaderboard":"/sota/multimodal-machine-translation-on-multi30k","task":"Multimodal Machine Translation","dataset":"Multi30K","model":"PS-KD","rank_in_archive_order":13,"of":15,"metrics":{"BLUE (DE-EN)":"32.3"},"uses_additional_data":false},{"leaderboard":"/sota/object-detection-on-pascal-voc-2007","task":"Object Detection","dataset":"PASCAL VOC 2007","model":"PS-KD (ResNet-152, CutMix)","rank_in_archive_order":12,"of":30,"metrics":{"MAP":"79.7%"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2006.12000","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2006.12000"}},"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/lgcnsai/ps-kd-pytorch","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":4,"unverified":1},"by_repo_kind":{"official":{"samples":5,"ran":4,"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":"d9def42110729a85","entry":"conv1x1","repo":"lgcnsai/ps-kd-pytorch","repo_kind":"official","path":"models/preact_resnet.py","file_url":"https://github.com/lgcnsai/ps-kd-pytorch/blob/HEAD/models/preact_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":"d9def42110729a85"}},{"code_sha256_prefix":"fac5364e2f53c6db","entry":"conv3x3","repo":"lgcnsai/ps-kd-pytorch","repo_kind":"official","path":"models/pyramid.py","file_url":"https://github.com/lgcnsai/ps-kd-pytorch/blob/HEAD/models/pyramid.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":"e18ed73aa89f051a","entry":"conv3x3","repo":"lgcnsai/ps-kd-pytorch","repo_kind":"official","path":"models/preact_resnet.py","file_url":"https://github.com/lgcnsai/ps-kd-pytorch/blob/HEAD/models/preact_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":"e18ed73aa89f051a"}},{"code_sha256_prefix":"03a0be4fb7381bd2","entry":"conv3x3","repo":"lgcnsai/ps-kd-pytorch","repo_kind":"official","path":"models/pyramid_shake_drop.py","file_url":"https://github.com/lgcnsai/ps-kd-pytorch/blob/HEAD/models/pyramid_shake_drop.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":"03a0be4fb7381bd2"}},{"code_sha256_prefix":"4530768e55e5fa0c","entry":"CIFAR_DenseNet121","repo":"lgcnsai/ps-kd-pytorch","repo_kind":"official","path":"models/densenet_cifar.py","file_url":"https://github.com/lgcnsai/ps-kd-pytorch/blob/HEAD/models/densenet_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":"4530768e55e5fa0c"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}