{"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/triple-generative-adversarial-nets","title":"Triple Generative Adversarial Nets","arxiv_id":"1703.02291","date":"2017-03-07","proceeding":"NeurIPS 2017 12","authors":["Chongxuan Li","Kun Xu","Jun Zhu","Bo Zhang"],"abstract":"Generative Adversarial Nets (GANs) have shown promise in image generation and\nsemi-supervised learning (SSL). However, existing GANs in SSL have two\nproblems: (1) the generator and the discriminator (i.e. the classifier) may not\nbe optimal at the same time; and (2) the generator cannot control the semantics\nof the generated samples. The problems essentially arise from the two-player\nformulation, where a single discriminator shares incompatible roles of\nidentifying fake samples and predicting labels and it only estimates the data\nwithout considering the labels. To address the problems, we present triple\ngenerative adversarial net (Triple-GAN), which consists of three players---a\ngenerator, a discriminator and a classifier. The generator and the classifier\ncharacterize the conditional distributions between images and labels, and the\ndiscriminator solely focuses on identifying fake image-label pairs. We design\ncompatible utilities to ensure that the distributions characterized by the\nclassifier and the generator both converge to the data distribution. Our\nresults on various datasets demonstrate that Triple-GAN as a unified model can\nsimultaneously (1) achieve the state-of-the-art classification results among\ndeep generative models, and (2) disentangle the classes and styles of the input\nand transfer smoothly in the data space via interpolation in the latent space\nclass-conditionally.","url_abs":"http://arxiv.org/abs/1703.02291v4","url_pdf":"http://arxiv.org/pdf/1703.02291v4.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":"triple-generative-adversarial-nets","repo_url":"https://github.com/zhenxuan00/triple-gan","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}}],"tasks":[{"task_slug":"image-generation","task_name":"Image Generation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=1703.02291","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}