{"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/policy-optimization-in-a-noisy-neighborhood-1","title":"Policy Optimization in a Noisy Neighborhood: On Return Landscapes in Continuous Control","arxiv_id":"2309.14597","date":"2023-09-26","proceeding":"NeurIPS 2023 11","authors":["Nate Rahn","Pierluca D'Oro","Harley Wiltzer","Pierre-Luc Bacon","Marc G. Bellemare"],"abstract":"Deep reinforcement learning agents for continuous control are known to exhibit significant instability in their performance over time. In this work, we provide a fresh perspective on these behaviors by studying the return landscape: the mapping between a policy and a return. We find that popular algorithms traverse noisy neighborhoods of this landscape, in which a single update to the policy parameters leads to a wide range of returns. By taking a distributional view of these returns, we map the landscape, characterizing failure-prone regions of policy space and revealing a hidden dimension of policy quality. We show that the landscape exhibits surprising structure by finding simple paths in parameter space which improve the stability of a policy. To conclude, we develop a distribution-aware procedure which finds such paths, navigating away from noisy neighborhoods in order to improve the robustness of a policy. Taken together, our results provide new insight into the optimization, evaluation, and design of agents.","url_abs":"https://arxiv.org/abs/2309.14597v3","url_pdf":"https://arxiv.org/pdf/2309.14597v3.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":"policy-optimization-in-a-noisy-neighborhood-1","repo_url":"https://github.com/nathanrahn/return-landscapes","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"jax","reach":{"status":"ok"}}],"tasks":[{"task_slug":"continuous-control","task_name":"Continuous Control"},{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"continuous-control","task_name":"continuous-control"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2309.14597","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2309.14597"}},"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/nathanrahn/return-landscapes","reach":{"status":"ok"}}],"summary":{"ran":6,"ran_draft_wrong":1,"unverified":2},"by_repo_kind":{"official":{"samples":9,"ran":7,"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":9,"samples":[{"code_sha256_prefix":"dd9c70cd9fdd268c","entry":"flatten","repo":"nathanrahn/return-landscapes","repo_kind":"official","path":"run_algorithm/util/dict_util.py","file_url":"https://github.com/nathanrahn/return-landscapes/blob/HEAD/run_algorithm/util/dict_util.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"dd9c70cd9fdd268c"}},{"code_sha256_prefix":"beb42d440ccf2925","entry":"layer_init","repo":"nathanrahn/return-landscapes","repo_kind":"official","path":"run_algorithm/experiments/ppo_atari_envpool.py","file_url":"https://github.com/nathanrahn/return-landscapes/blob/HEAD/run_algorithm/experiments/ppo_atari_envpool.py","link_basis":"harvester_set","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"beb42d440ccf2925"}},{"code_sha256_prefix":"c3b19b121ec34ed3","entry":"maybe_parse_reset_bounds","repo":"nathanrahn/return-landscapes","repo_kind":"official","path":"run_algorithm/experiments/utils.py","file_url":"https://github.com/nathanrahn/return-landscapes/blob/HEAD/run_algorithm/experiments/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c3b19b121ec34ed3"}},{"code_sha256_prefix":"69d25d09f8b903bf","entry":"prefix_dict_keys","repo":"nathanrahn/return-landscapes","repo_kind":"official","path":"run_algorithm/experiments/train_bc.py","file_url":"https://github.com/nathanrahn/return-landscapes/blob/HEAD/run_algorithm/experiments/train_bc.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"69d25d09f8b903bf"}},{"code_sha256_prefix":"6167c00d6c136928","entry":"unflatten","repo":"nathanrahn/return-landscapes","repo_kind":"official","path":"run_algorithm/util/dict_util.py","file_url":"https://github.com/nathanrahn/return-landscapes/blob/HEAD/run_algorithm/util/dict_util.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"6167c00d6c136928"}},{"code_sha256_prefix":"ba83c9745d123775","entry":"verify_number_and_cast","repo":"nathanrahn/return-landscapes","repo_kind":"official","path":"run_algorithm/experiments/utils.py","file_url":"https://github.com/nathanrahn/return-landscapes/blob/HEAD/run_algorithm/experiments/utils.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ba83c9745d123775"}},{"code_sha256_prefix":"d1d15d620d075bbe","entry":"which_cluster","repo":"nathanrahn/return-landscapes","repo_kind":"official","path":"run_algorithm/environment.py","file_url":"https://github.com/nathanrahn/return-landscapes/blob/HEAD/run_algorithm/environment.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"d1d15d620d075bbe"}},{"code_sha256_prefix":"0bd4598ee260c0c4","entry":"copy_model","repo":"nathanrahn/return-landscapes","repo_kind":"official","path":"run_algorithm/experiments/utils.py","file_url":"https://github.com/nathanrahn/return-landscapes/blob/HEAD/run_algorithm/experiments/utils.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"0bd4598ee260c0c4"}},{"code_sha256_prefix":"a993dda29208abe3","entry":"get_run_by_run_name","repo":"nathanrahn/return-landscapes","repo_kind":"official","path":"improve_ltp/util/load_models.py","file_url":"https://github.com/nathanrahn/return-landscapes/blob/HEAD/improve_ltp/util/load_models.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a993dda29208abe3"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}