{"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/local-policy-search-with-bayesian","title":"Local policy search with Bayesian optimization","arxiv_id":"2106.11899","date":"2021-06-22","proceeding":"NeurIPS 2021 12","authors":["Sarah Müller","Alexander von Rohr","Sebastian Trimpe"],"abstract":"Reinforcement learning (RL) aims to find an optimal policy by interaction with an environment. Consequently, learning complex behavior requires a vast number of samples, which can be prohibitive in practice. Nevertheless, instead of systematically reasoning and actively choosing informative samples, policy gradients for local search are often obtained from random perturbations. These random samples yield high variance estimates and hence are sub-optimal in terms of sample complexity. Actively selecting informative samples is at the core of Bayesian optimization, which constructs a probabilistic surrogate of the objective from past samples to reason about informative subsequent ones. In this paper, we propose to join both worlds. We develop an algorithm utilizing a probabilistic model of the objective function and its gradient. Based on the model, the algorithm decides where to query a noisy zeroth-order oracle to improve the gradient estimates. The resulting algorithm is a novel type of policy search method, which we compare to existing black-box algorithms. The comparison reveals improved sample complexity and reduced variance in extensive empirical evaluations on synthetic objectives. Further, we highlight the benefits of active sampling on popular RL benchmarks.","url_abs":"https://arxiv.org/abs/2106.11899v2","url_pdf":"https://arxiv.org/pdf/2106.11899v2.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":"local-policy-search-with-bayesian","repo_url":"https://github.com/sarmueller/gibo","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}},{"paper_slug":"local-policy-search-with-bayesian","repo_url":"https://github.com/PaulsonLab/CAGES","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok"}},{"paper_slug":"local-policy-search-with-bayesian","repo_url":"https://github.com/kayween/local-bo-mpd","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"bayesian-optimization","task_name":"Bayesian Optimization"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2106.11899","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2106.11899"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-24T18:15:14+00:00","read_at_is":"when the build read Syntology's graph, not when any sample ran","claim":"Per-sample execution status on synthesized fixtures; not a correctness claim about the paper. Samples come from repositories linked to the paper, official or community; repo_kind says which.","repos":[{"provenance":"deterministic:regex_extraction","url":"https://github.com/sarmueller/gibo","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/kayween/local-bo-mpd","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/PaulsonLab/CAGES","reach":{"status":"ok"}}],"summary":{"unverified":8},"by_repo_kind":{"official":{"samples":8,"ran":0,"repositories":1}},"repo_kind_vocabulary":{"official":"The archive marks this repository official for the paper","named_in_paper":"The archive records that the paper mentions this repository; it is not marked official","listed":"In the archive's code links for this paper, not marked official and not recorded as mentioned in the paper","found_in_text":"Syntology found this repository in the paper's own text; whether it is the authors' implementation is not asserted","community":"Not in the archive's code links for this paper; a community repository Syntology harvested"},"n_pointer_only_for_licence":0,"samples":[{"code_sha256_prefix":"5a80e2f4f753b250","entry":"call_counter","repo":"sarmueller/gibo","repo_kind":"official","path":"src/loop.py","file_url":"https://github.com/sarmueller/gibo/blob/HEAD/src/loop.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"5a80e2f4f753b250"}},{"code_sha256_prefix":"9ba1bb4104c6cbdc","entry":"insert","repo":"sarmueller/gibo","repo_kind":"official","path":"src/config.py","file_url":"https://github.com/sarmueller/gibo/blob/HEAD/src/config.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9ba1bb4104c6cbdc"}},{"code_sha256_prefix":"4413d8e3286e6243","entry":"manipulate_reward","repo":"sarmueller/gibo","repo_kind":"official","path":"src/environment_api.py","file_url":"https://github.com/sarmueller/gibo/blob/HEAD/src/environment_api.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4413d8e3286e6243"}},{"code_sha256_prefix":"df94ac83c17a7fb3","entry":"one_step_cholesky","repo":"sarmueller/gibo","repo_kind":"official","path":"src/cholesky.py","file_url":"https://github.com/sarmueller/gibo/blob/HEAD/src/cholesky.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"df94ac83c17a7fb3"}},{"code_sha256_prefix":"20307dfa91e06968","entry":"optimize_acqf_custom_bo","repo":"sarmueller/gibo","repo_kind":"official","path":"src/acquisition_function.py","file_url":"https://github.com/sarmueller/gibo/blob/HEAD/src/acquisition_function.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"20307dfa91e06968"}},{"code_sha256_prefix":"bf9e0cbb23842154","entry":"optimize_acqf_vanilla_bo","repo":"sarmueller/gibo","repo_kind":"official","path":"src/acquisition_function.py","file_url":"https://github.com/sarmueller/gibo/blob/HEAD/src/acquisition_function.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"bf9e0cbb23842154"}},{"code_sha256_prefix":"93332fe605a1eec4","entry":"postprocessing_interpolate_x","repo":"sarmueller/gibo","repo_kind":"official","path":"src/evaluate.py","file_url":"https://github.com/sarmueller/gibo/blob/HEAD/src/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"93332fe605a1eec4"}},{"code_sha256_prefix":"9493a6a3cdcb4b44","entry":"sort_rewards_global_optimization","repo":"sarmueller/gibo","repo_kind":"official","path":"src/evaluate.py","file_url":"https://github.com/sarmueller/gibo/blob/HEAD/src/evaluate.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9493a6a3cdcb4b44"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}