{"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/improving-gan-training-with-probability-ratio","title":"Improving GAN Training with Probability Ratio Clipping and Sample Reweighting","arxiv_id":"2006.06900","date":"2020-06-12","proceeding":"NeurIPS 2020 12","authors":["Yue Wu","Pan Zhou","Andrew Gordon Wilson","Eric P. Xing","Zhiting Hu"],"abstract":"Despite success on a wide range of problems related to vision, generative adversarial networks (GANs) often suffer from inferior performance due to unstable training, especially for text generation. To solve this issue, we propose a new variational GAN training framework which enjoys superior training stability. Our approach is inspired by a connection of GANs and reinforcement learning under a variational perspective. The connection leads to (1) probability ratio clipping that regularizes generator training to prevent excessively large updates, and (2) a sample re-weighting mechanism that improves discriminator training by downplaying bad-quality fake samples. Moreover, our variational GAN framework can provably overcome the training issue in many GANs that an optimal discriminator cannot provide any informative gradient to training generator. By plugging the training approach in diverse state-of-the-art GAN architectures, we obtain significantly improved performance over a range of tasks, including text generation, text style transfer, and image generation.","url_abs":"https://arxiv.org/abs/2006.06900v4","url_pdf":"https://arxiv.org/pdf/2006.06900v4.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":"improving-gan-training-with-probability-ratio","repo_url":"https://github.com/Holmeswww/PPOGAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"},{"task_slug":"style-transfer","task_name":"Style Transfer"},{"task_slug":"text-generation","task_name":"Text Generation"},{"task_slug":"text-style-transfoer","task_name":"Text Style Transfer"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-generation-on-cifar-10","task":"Image Generation","dataset":"CIFAR-10","model":"PPOGAN","rank_in_archive_order":45,"of":78,"metrics":{"FID":"10.7"},"uses_additional_data":false},{"leaderboard":"/sota/text-generation-on-emnlp2017-wmt","task":"Text Generation","dataset":"EMNLP2017 WMT","model":"PPOGAN","rank_in_archive_order":2,"of":5,"metrics":{"BLEU-2":"0.905","BLEU-3":"0.692","BLEU-4":"0.47","BLEU-5":"0.322","NLLgen":"2.265"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2006.06900","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2006.06900"}},"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/Holmeswww/PPOGAN","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"unverified":3},"by_repo_kind":{"official":{"samples":3,"ran":0,"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":"0def50a351111624","entry":"calculate_frechet_distance","repo":"Holmeswww/PPOGAN","repo_kind":"official","path":"cifar10/fid-score.py","file_url":"https://github.com/Holmeswww/PPOGAN/blob/HEAD/cifar10/fid-score.py","link_basis":"harvester_set","language":"python","status":"unverified","verification_level":0,"contract_check":"RAISES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"0def50a351111624"}},{"code_sha256_prefix":"d8ab4cdbe05b528e","entry":"clean_text","repo":"Holmeswww/PPOGAN","repo_kind":"official","path":"style_transfer/prepare_manual.py","file_url":"https://github.com/Holmeswww/PPOGAN/blob/HEAD/style_transfer/prepare_manual.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":"d8ab4cdbe05b528e"}},{"code_sha256_prefix":"6bcfa8379bf537c5","entry":"dynamic_padding","repo":"Holmeswww/PPOGAN","repo_kind":"official","path":"style_transfer/ctrl_gen_model.py","file_url":"https://github.com/Holmeswww/PPOGAN/blob/HEAD/style_transfer/ctrl_gen_model.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":"6bcfa8379bf537c5"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}