{"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/knowledge-distillation-with-adversarial","title":"Knowledge Distillation with Adversarial Samples Supporting Decision Boundary","arxiv_id":"1805.05532","date":"2018-05-15","proceeding":null,"authors":["Byeongho Heo","Minsik Lee","Sangdoo Yun","Jin Young Choi"],"abstract":"Many recent works on knowledge distillation have provided ways to transfer\nthe knowledge of a trained network for improving the learning process of a new\none, but finding a good technique for knowledge distillation is still an open\nproblem. In this paper, we provide a new perspective based on a decision\nboundary, which is one of the most important component of a classifier. The\ngeneralization performance of a classifier is closely related to the adequacy\nof its decision boundary, so a good classifier bears a good decision boundary.\nTherefore, transferring information closely related to the decision boundary\ncan be a good attempt for knowledge distillation. To realize this goal, we\nutilize an adversarial attack to discover samples supporting a decision\nboundary. Based on this idea, to transfer more accurate information about the\ndecision boundary, the proposed algorithm trains a student classifier based on\nthe adversarial samples supporting the decision boundary. Experiments show that\nthe proposed method indeed improves knowledge distillation and achieves the\nstate-of-the-arts performance.","url_abs":"http://arxiv.org/abs/1805.05532v4","url_pdf":"http://arxiv.org/pdf/1805.05532v4.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":"knowledge-distillation-with-adversarial","repo_url":"https://github.com/bhheo/BSS_distillation","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"adversarial-attack","task_name":"Adversarial Attack"},{"task_slug":"knowledge-distillation","task_name":"Knowledge Distillation"}],"methods":[{"method_slug":"knowledge-distillation","method_name":"Knowledge Distillation"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1805.05532","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1805.05532"}},"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/bhheo/BSS_distillation","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":1,"unverified":3},"by_repo_kind":{"official":{"samples":4,"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":"adf70cc0c6fe2f98","entry":"reduce_sum","repo":"bhheo/BSS_distillation","repo_kind":"official","path":"attacks/helpers.py","file_url":"https://github.com/bhheo/BSS_distillation/blob/HEAD/attacks/helpers.py","link_basis":"plan_row","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":"adf70cc0c6fe2f98"}},{"code_sha256_prefix":"f5b74e1e9898886a","entry":"BN_version_fix","repo":"bhheo/BSS_distillation","repo_kind":"official","path":"models/resnet.py","file_url":"https://github.com/bhheo/BSS_distillation/blob/HEAD/models/resnet.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":"f5b74e1e9898886a"}},{"code_sha256_prefix":"3f6d01fa8467254a","entry":"reduce_mean","repo":"bhheo/BSS_distillation","repo_kind":"official","path":"attacks/helpers.py","file_url":"https://github.com/bhheo/BSS_distillation/blob/HEAD/attacks/helpers.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":"3f6d01fa8467254a"}},{"code_sha256_prefix":"4d44936ca9dfd048","entry":"reduce_min","repo":"bhheo/BSS_distillation","repo_kind":"official","path":"attacks/helpers.py","file_url":"https://github.com/bhheo/BSS_distillation/blob/HEAD/attacks/helpers.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":"4d44936ca9dfd048"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}