{"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/implementation-of-stochastic-quasi-newtons","title":"Implementation of Stochastic Quasi-Newton's Method in PyTorch","arxiv_id":"1805.02338","date":"2018-05-07","proceeding":null,"authors":["Yingkai Li","Huidong Liu"],"abstract":"In this paper, we implement the Stochastic Damped LBFGS (SdLBFGS) for\nstochastic non-convex optimization. We make two important modifications to the\noriginal SdLBFGS algorithm. First, by initializing the Hessian at each step\nusing an identity matrix, the algorithm converges better than original\nalgorithm. Second, by performing direction normalization we could gain stable\noptimization procedure without line search. Experiments on minimizing a 2D\nnon-convex function shows that our improved algorithm converges better than\noriginal algorithm, and experiments on the CIFAR10 and MNIST datasets show that\nour improved algorithm works stably and gives comparable or even better testing\naccuracies than first order optimizers SGD, Adagrad, and second order\noptimizers LBFGS in PyTorch.","url_abs":"http://arxiv.org/abs/1805.02338v1","url_pdf":"http://arxiv.org/pdf/1805.02338v1.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":"implementation-of-stochastic-quasi-newtons","repo_url":"https://github.com/harryliew/SdLBFGS","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":null}],"tasks":[],"methods":[{"method_slug":"sgd","method_name":"SGD"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1805.02338","atlas_url":"https://app.syntology.ai/?focus=1805.02338","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}