{"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-mmd-gan-training-with-repulsive","title":"Improving MMD-GAN Training with Repulsive Loss Function","arxiv_id":"1812.09916","date":"2018-12-24","proceeding":"ICLR 2019 5","authors":["Wei Wang","Yuan Sun","Saman Halgamuge"],"abstract":"Generative adversarial nets (GANs) are widely used to learn the data sampling\nprocess and their performance may heavily depend on the loss functions, given a\nlimited computational budget. This study revisits MMD-GAN that uses the maximum\nmean discrepancy (MMD) as the loss function for GAN and makes two\ncontributions. First, we argue that the existing MMD loss function may\ndiscourage the learning of fine details in data as it attempts to contract the\ndiscriminator outputs of real data. To address this issue, we propose a\nrepulsive loss function to actively learn the difference among the real data by\nsimply rearranging the terms in MMD. Second, inspired by the hinge loss, we\npropose a bounded Gaussian kernel to stabilize the training of MMD-GAN with the\nrepulsive loss function. The proposed methods are applied to the unsupervised\nimage generation tasks on CIFAR-10, STL-10, CelebA, and LSUN bedroom datasets.\nResults show that the repulsive loss function significantly improves over the\nMMD loss at no additional computational cost and outperforms other\nrepresentative loss functions. The proposed methods achieve an FID score of\n16.21 on the CIFAR-10 dataset using a single DCGAN network and spectral\nnormalization.","url_abs":"http://arxiv.org/abs/1812.09916v4","url_pdf":"http://arxiv.org/pdf/1812.09916v4.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-mmd-gan-training-with-repulsive","repo_url":"https://github.com/richardwth/MMD-GAN","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"convolution","method_name":"Convolution"},{"method_slug":"dcgan","method_name":"DCGAN"},{"method_slug":"relu","method_name":"ReLU"}],"datasets_introduced":[],"methods_introduced":[],"results":[{"leaderboard":"/sota/image-generation-on-stl-10","task":"Image Generation","dataset":"STL-10","model":"Improving MMD GAN","rank_in_archive_order":25,"of":31,"metrics":{"FID":"37.63","Inception score":"9.34"},"uses_additional_data":false}],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1812.09916","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"1812.09916"}},"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/richardwth/MMD-GAN","reach":{"status":"ok","spdx":"Apache-2.0"}}],"summary":{"unverified":9},"by_repo_kind":{"official":{"samples":9,"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":"e255f546eb51ae30","entry":"average_tower_grads","repo":"richardwth/MMD-GAN","repo_kind":"official","path":"GeneralTools/graph_func.py","file_url":"https://github.com/richardwth/MMD-GAN/blob/HEAD/GeneralTools/graph_func.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e255f546eb51ae30"}},{"code_sha256_prefix":"18fcbff51eccaf48","entry":"bias_initializer","repo":"richardwth/MMD-GAN","repo_kind":"official","path":"GeneralTools/layer_func.py","file_url":"https://github.com/richardwth/MMD-GAN/blob/HEAD/GeneralTools/layer_func.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"18fcbff51eccaf48"}},{"code_sha256_prefix":"3c3e116e015d542b","entry":"create_variable","repo":"richardwth/MMD-GAN","repo_kind":"official","path":"GeneralTools/graph_func.py","file_url":"https://github.com/richardwth/MMD-GAN/blob/HEAD/GeneralTools/graph_func.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"3c3e116e015d542b"}},{"code_sha256_prefix":"9f0f284ad1588861","entry":"get_files_in_child_folders","repo":"richardwth/MMD-GAN","repo_kind":"official","path":"GeneralTools/input_func.py","file_url":"https://github.com/richardwth/MMD-GAN/blob/HEAD/GeneralTools/input_func.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"9f0f284ad1588861"}},{"code_sha256_prefix":"47bf2bd6f4c07778","entry":"kron_by_reshape","repo":"richardwth/MMD-GAN","repo_kind":"official","path":"GeneralTools/math_func.py","file_url":"https://github.com/richardwth/MMD-GAN/blob/HEAD/GeneralTools/math_func.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"47bf2bd6f4c07778"}},{"code_sha256_prefix":"37a70070238648d3","entry":"mean_cov_np","repo":"richardwth/MMD-GAN","repo_kind":"official","path":"GeneralTools/math_func.py","file_url":"https://github.com/richardwth/MMD-GAN/blob/HEAD/GeneralTools/math_func.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"37a70070238648d3"}},{"code_sha256_prefix":"57976e3d8ed9ad0a","entry":"raw_image_to_np","repo":"richardwth/MMD-GAN","repo_kind":"official","path":"GeneralTools/input_func.py","file_url":"https://github.com/richardwth/MMD-GAN/blob/HEAD/GeneralTools/input_func.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"57976e3d8ed9ad0a"}},{"code_sha256_prefix":"c3836b2356ce9774","entry":"scale_range","repo":"richardwth/MMD-GAN","repo_kind":"official","path":"GeneralTools/math_func.py","file_url":"https://github.com/richardwth/MMD-GAN/blob/HEAD/GeneralTools/math_func.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c3836b2356ce9774"}},{"code_sha256_prefix":"b38f1ddd69e392d9","entry":"xavier_init","repo":"richardwth/MMD-GAN","repo_kind":"official","path":"GeneralTools/graph_func.py","file_url":"https://github.com/richardwth/MMD-GAN/blob/HEAD/GeneralTools/graph_func.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"b38f1ddd69e392d9"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}