{"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/no-modes-left-behind-capturing-the-data","title":"No Modes left behind: Capturing the data distribution effectively using GANs","arxiv_id":"1802.00771","date":"2018-02-02","proceeding":null,"authors":["Shashank Sharma","Vinay P. Namboodiri"],"abstract":"Generative adversarial networks (GANs) while being very versatile in\nrealistic image synthesis, still are sensitive to the input distribution. Given\na set of data that has an imbalance in the distribution, the networks are\nsusceptible to missing modes and not capturing the data distribution. While\nvarious methods have been tried to improve training of GANs, these have not\naddressed the challenges of covering the full data distribution. Specifically,\na generator is not penalized for missing a mode. We show that these are\ntherefore still susceptible to not capturing the full data distribution.\n  In this paper, we propose a simple approach that combines an encoder based\nobjective with novel loss functions for generator and discriminator that\nimproves the solution in terms of capturing missing modes. We validate that the\nproposed method results in substantial improvements through its detailed\nanalysis on toy and real datasets. The quantitative and qualitative results\ndemonstrate that the proposed method improves the solution for the problem of\nmissing modes and improves training of GANs.","url_abs":"http://arxiv.org/abs/1802.00771v1","url_pdf":"http://arxiv.org/pdf/1802.00771v1.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":"no-modes-left-behind-capturing-the-data","repo_url":"https://github.com/shashank879/logan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"tf","reach":null}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}