{"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/generalizing-hamiltonian-monte-carlo-with","title":"Generalizing Hamiltonian Monte Carlo with Neural Networks","arxiv_id":"1711.09268","date":"2017-11-25","proceeding":"ICLR 2018 1","authors":["Daniel Levy","Matthew D. Hoffman","Jascha Sohl-Dickstein"],"abstract":"We present a general-purpose method to train Markov chain Monte Carlo\nkernels, parameterized by deep neural networks, that converge and mix quickly\nto their target distribution. Our method generalizes Hamiltonian Monte Carlo\nand is trained to maximize expected squared jumped distance, a proxy for mixing\nspeed. We demonstrate large empirical gains on a collection of simple but\nchallenging distributions, for instance achieving a 106x improvement in\neffective sample size in one case, and mixing when standard HMC makes no\nmeasurable progress in a second. Finally, we show quantitative and qualitative\ngains on a real-world task: latent-variable generative modeling. We release an\nopen source TensorFlow implementation of the algorithm.","url_abs":"http://arxiv.org/abs/1711.09268v3","url_pdf":"http://arxiv.org/pdf/1711.09268v3.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":"generalizing-hamiltonian-monte-carlo-with","repo_url":"https://github.com/brain-research/l2hmc","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":{"status":"unanswered"}},{"paper_slug":"generalizing-hamiltonian-monte-carlo-with","repo_url":"https://github.com/saforem2/l2hmc-qcd","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":{"status":"ok","spdx":"Apache-2.0"}},{"paper_slug":"generalizing-hamiltonian-monte-carlo-with","repo_url":"https://github.com/soran-ghaderi/torchebm","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"unanswered"}}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/1711.09268","atlas_url":"https://app.syntology.ai/?focus=1711.09268","mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}