{"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/noisy-blackbox-optimization-with-multi","title":"Noisy Blackbox Optimization with Multi-Fidelity Queries: A Tree Search Approach","arxiv_id":"1810.10482","date":"2018-10-24","proceeding":null,"authors":["Rajat Sen","Kirthevasan Kandasamy","Sanjay Shakkottai"],"abstract":"We study the problem of black-box optimization of a noisy function in the\npresence of low-cost approximations or fidelities, which is motivated by\nproblems like hyper-parameter tuning. In hyper-parameter tuning evaluating the\nblack-box function at a point involves training a learning algorithm on a large\ndata-set at a particular hyper-parameter and evaluating the validation error.\nEven a single such evaluation can be prohibitively expensive. Therefore, it is\nbeneficial to use low-cost approximations, like training the learning algorithm\non a sub-sampled version of the whole data-set. These low-cost\napproximations/fidelities can however provide a biased and noisy estimate of\nthe function value. In this work, we incorporate the multi-fidelity setup in\nthe powerful framework of noisy black-box optimization through tree-like\nhierarchical partitions. We propose a multi-fidelity bandit based tree-search\nalgorithm for the problem and provide simple regret bounds for our algorithm.\nFinally, we validate the performance of our algorithm on real and synthetic\ndatasets, where it outperforms several benchmarks.","url_abs":"http://arxiv.org/abs/1810.10482v1","url_pdf":"http://arxiv.org/pdf/1810.10482v1.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":"noisy-blackbox-optimization-with-multi","repo_url":"https://github.com/rajatsen91/MFTreeSearchCV","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"none","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":null,"mcp":null,"developers":"https://syntology.ai/developers"},"arxiv_metadata":null,"syntology_extracted_results":null}