{"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/connections-between-support-vector-machines","title":"Gradient penalty from a maximum margin perspective","arxiv_id":"1910.06922","date":"2019-10-15","proceeding":null,"authors":["Alexia Jolicoeur-Martineau","Ioannis Mitliagkas"],"abstract":"A popular heuristic for improved performance in Generative adversarial networks (GANs) is to use some form of gradient penalty on the discriminator. This gradient penalty was originally motivated by a Wasserstein distance formulation. However, the use of gradient penalty in other GAN formulations is not well motivated. We present a unifying framework of expected margin maximization and show that a wide range of gradient-penalized GANs (e.g., Wasserstein, Standard, Least-Squares, and Hinge GANs) can be derived from this framework. Our results imply that employing gradient penalties induces a large-margin classifier (thus, a large-margin discriminator in GANs). We describe how expected margin maximization helps reduce vanishing gradients at fake (generated) samples, a known problem in GANs. From this framework, we derive a new $L^\\infty$ gradient norm penalty with Hinge loss which generally produces equally good (or better) generated output in GANs than $L^2$-norm penalties (based on the Fr\\'echet Inception Distance).","url_abs":"https://arxiv.org/abs/1910.06922v2","url_pdf":"https://arxiv.org/pdf/1910.06922v2.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":"connections-between-support-vector-machines","repo_url":"https://github.com/AlexiaJM/MaximumMarginGANs","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"connections-between-support-vector-machines","repo_url":"https://github.com/lucidrains/stylegan2-pytorch","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[{"method_slug":"convolution","method_name":"Convolution"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-generation-on-cifar-10","task":"Image Generation","dataset":"CIFAR-10","model":"HingeGAN","rank_in_archive_order":65,"of":78,"metrics":{"FID":"27.12"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1910.06922","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1910.06922"}},"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/lucidrains/stylegan2-pytorch","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/AlexiaJM/MaximumMarginGANs","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_fixture":1,"ran_violates":1,"unverified":8},"by_repo_kind":{"official":{"samples":10,"ran":2,"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":"7e61fe013d57c481","entry":"calculate_frechet_distance","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/fid.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/fid.py","link_basis":"plan_row","language":"python","status":"ran_fixture","verification_level":1,"contract_check":"RAISES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"7e61fe013d57c481"}},{"code_sha256_prefix":"9617b12aa6ac6bbe","entry":"strToBool","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/GAN.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/GAN.py","link_basis":"plan_row","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9617b12aa6ac6bbe"}},{"code_sha256_prefix":"6917ffcb1214d74f","entry":"calculate_activation_statistics","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/fid.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/fid.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":"6917ffcb1214d74f"}},{"code_sha256_prefix":"846d824d45774de9","entry":"calculate_frechet_distance","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/fid_new.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/fid_new.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":"846d824d45774de9"}},{"code_sha256_prefix":"e69fcd6206886931","entry":"get_activations","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/fid.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/fid.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":"e69fcd6206886931"}},{"code_sha256_prefix":"52a3f79b47513c42","entry":"make_dot","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/pytorch_visualize.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/pytorch_visualize.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":"52a3f79b47513c42"}},{"code_sha256_prefix":"7376fdf98a6f6485","entry":"preprocessCatFace","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/preprocess_cat_dataset.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/preprocess_cat_dataset.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":"7376fdf98a6f6485"}},{"code_sha256_prefix":"c3ee865ac3297b5a","entry":"register_vis_hooks","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/pytorch_visualize.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/pytorch_visualize.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":"c3ee865ac3297b5a"}},{"code_sha256_prefix":"cf539e0bbc356efb","entry":"rotateCoords","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/preprocess_cat_dataset.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/preprocess_cat_dataset.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":"cf539e0bbc356efb"}},{"code_sha256_prefix":"17dcbb4211f84f3f","entry":"save_visualization","repo":"AlexiaJM/MaximumMarginGANs","repo_kind":"official","path":"Code/pytorch_visualize.py","file_url":"https://github.com/AlexiaJM/MaximumMarginGANs/blob/HEAD/Code/pytorch_visualize.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":"17dcbb4211f84f3f"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}