{"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/iclr-reproducibility-challenge-report-padam","title":"ICLR Reproducibility Challenge Report (Padam : Closing The Generalization Gap Of Adaptive Gradient Methods in Training Deep Neural Networks)","arxiv_id":"1901.09517","date":"2019-01-28","proceeding":null,"authors":["Harshal Mittal","Kartikey Pandey","Yash Kant"],"abstract":"This work is a part of ICLR Reproducibility Challenge 2019, we try to\nreproduce the results in the conference submission PADAM: Closing The\nGeneralization Gap of Adaptive Gradient Methods In Training Deep Neural\nNetworks. Adaptive gradient methods proposed in past demonstrate a degraded\ngeneralization performance than the stochastic gradient descent (SGD) with\nmomentum. The authors try to address this problem by designing a new\noptimization algorithm that bridges the gap between the space of Adaptive\nGradient algorithms and SGD with momentum. With this method a new tunable\nhyperparameter called partially adaptive parameter p is introduced that varies\nbetween [0, 0.5]. We build the proposed optimizer and use it to mirror the\nexperiments performed by the authors. We review and comment on the empirical\nanalysis performed by the authors. Finally, we also propose a future direction\nfor further study of Padam. Our code is available at:\nhttps://github.com/yashkant/Padam-Tensorflow","url_abs":"http://arxiv.org/abs/1901.09517v1","url_pdf":"http://arxiv.org/pdf/1901.09517v1.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":"iclr-reproducibility-challenge-report-padam","repo_url":"https://github.com/yashkant/Padam-Tensorflow","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[{"method_slug":"sgd","method_name":"SGD"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":null,"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}