{"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/bigger-regularized-optimistic-scaling-for","title":"Bigger, Regularized, Optimistic: scaling for compute and sample-efficient continuous control","arxiv_id":"2405.16158","date":"2024-05-25","proceeding":null,"authors":["Michal Nauman","Mateusz Ostaszewski","Krzysztof Jankowski","Piotr Miłoś","Marek Cygan"],"abstract":"Sample efficiency in Reinforcement Learning (RL) has traditionally been driven by algorithmic enhancements. In this work, we demonstrate that scaling can also lead to substantial improvements. We conduct a thorough investigation into the interplay of scaling model capacity and domain-specific RL enhancements. These empirical findings inform the design choices underlying our proposed BRO (Bigger, Regularized, Optimistic) algorithm. The key innovation behind BRO is that strong regularization allows for effective scaling of the critic networks, which, paired with optimistic exploration, leads to superior performance. BRO achieves state-of-the-art results, significantly outperforming the leading model-based and model-free algorithms across 40 complex tasks from the DeepMind Control, MetaWorld, and MyoSuite benchmarks. BRO is the first model-free algorithm to achieve near-optimal policies in the notoriously challenging Dog and Humanoid tasks.","url_abs":"https://arxiv.org/abs/2405.16158v3","url_pdf":"https://arxiv.org/pdf/2405.16158v3.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":"bigger-regularized-optimistic-scaling-for","repo_url":"https://github.com/naumix/BiggerRegularizedOptimistic","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"jax","reach":null}],"tasks":[{"task_slug":"continuous-control","task_name":"Continuous Control"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"continuous-control","task_name":"continuous-control"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2405.16158","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.16158"}},"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/sfujim/TD3","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/ikostrikov/jaxrl","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/naumix/BiggerRegularizedOptimistic","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/adityab/CrossQ","reach":{"status":"ok","spdx":"NOASSERTION"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/denisyarats/pytorch_sac","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/proceduralia/high_replay_ratio_continuous_control","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/nicklashansen/tdmpc2","reach":null}],"summary":{"ran":7,"ran_violates":1,"ran_draft_wrong":1,"ran_honours":2,"unverified":9},"by_repo_kind":{"official":{"samples":8,"ran":3,"repositories":1},"found_in_text":{"samples":12,"ran":8,"repositories":4}},"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":1,"samples":[{"code_sha256_prefix":"9c81de6b9c4c134f","entry":"Actor","repo":"sfujim/TD3","repo_kind":"found_in_text","path":"TD3.py","file_url":"https://github.com/sfujim/TD3/blob/HEAD/TD3.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9c81de6b9c4c134f"}},{"code_sha256_prefix":"78955e8aa0ecd20a","entry":"BatchRenorm","repo":"adityab/CrossQ","repo_kind":"found_in_text","path":"sbx/sac/policies.py","file_url":"https://github.com/adityab/CrossQ/blob/HEAD/sbx/sac/policies.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NOASSERTION","inline_ok":false,"mcp_get_code":{"code_sha256":"78955e8aa0ecd20a"}},{"code_sha256_prefix":"8e4b15accbb74070","entry":"Model","repo":"proceduralia/high_replay_ratio_continuous_control","repo_kind":"found_in_text","path":"jaxrl/agents/sac/sac_learner.py","file_url":"https://github.com/proceduralia/high_replay_ratio_continuous_control/blob/HEAD/jaxrl/agents/sac/sac_learner.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"8e4b15accbb74070"}},{"code_sha256_prefix":"60af6b78baad020d","entry":"Model","repo":"naumix/BiggerRegularizedOptimistic","repo_kind":"official","path":"jaxrl/bro/bro_learner.py","file_url":"https://github.com/naumix/BiggerRegularizedOptimistic/blob/HEAD/jaxrl/bro/bro_learner.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"60af6b78baad020d"}},{"code_sha256_prefix":"cfb82327f2389742","entry":"SaveState","repo":"proceduralia/high_replay_ratio_continuous_control","repo_kind":"found_in_text","path":"jaxrl/agents/sac/sac_learner.py","file_url":"https://github.com/proceduralia/high_replay_ratio_continuous_control/blob/HEAD/jaxrl/agents/sac/sac_learner.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"cfb82327f2389742"}},{"code_sha256_prefix":"aa4126690e869bbf","entry":"SaveState","repo":"naumix/BiggerRegularizedOptimistic","repo_kind":"official","path":"jaxrl/bro/bro_learner.py","file_url":"https://github.com/naumix/BiggerRegularizedOptimistic/blob/HEAD/jaxrl/bro/bro_learner.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"aa4126690e869bbf"}},{"code_sha256_prefix":"590f1bfec73e823b","entry":"TD3","repo":"sfujim/TD3","repo_kind":"found_in_text","path":"TD3.py","file_url":"https://github.com/sfujim/TD3/blob/HEAD/TD3.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"590f1bfec73e823b"}},{"code_sha256_prefix":"ae8a2984d9e381fd","entry":"huber_replace","repo":"naumix/BiggerRegularizedOptimistic","repo_kind":"official","path":"jaxrl/bro/bro_learner.py","file_url":"https://github.com/naumix/BiggerRegularizedOptimistic/blob/HEAD/jaxrl/bro/bro_learner.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"ae8a2984d9e381fd"}},{"code_sha256_prefix":"db00dda8b96201d3","entry":"mlp","repo":"denisyarats/pytorch_sac","repo_kind":"found_in_text","path":"utils.py","file_url":"https://github.com/denisyarats/pytorch_sac/blob/HEAD/utils.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"db00dda8b96201d3"}},{"code_sha256_prefix":"6ee8c9901ab984a9","entry":"to_np","repo":"denisyarats/pytorch_sac","repo_kind":"found_in_text","path":"utils.py","file_url":"https://github.com/denisyarats/pytorch_sac/blob/HEAD/utils.py","link_basis":"plan_row","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"6ee8c9901ab984a9"}},{"code_sha256_prefix":"dd6b087dfe582577","entry":"tree_norm","repo":"proceduralia/high_replay_ratio_continuous_control","repo_kind":"found_in_text","path":"jaxrl/agents/sac/sac_learner.py","file_url":"https://github.com/proceduralia/high_replay_ratio_continuous_control/blob/HEAD/jaxrl/agents/sac/sac_learner.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"well_formed","behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"dd6b087dfe582577"}},{"code_sha256_prefix":"ce00f81a9f2c68ec","entry":"BRO","repo":"naumix/BiggerRegularizedOptimistic","repo_kind":"official","path":"jaxrl/bro/bro_learner.py","file_url":"https://github.com/naumix/BiggerRegularizedOptimistic/blob/HEAD/jaxrl/bro/bro_learner.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":"ce00f81a9f2c68ec"}},{"code_sha256_prefix":"09d56aa25dec1123","entry":"SACLearner","repo":"proceduralia/high_replay_ratio_continuous_control","repo_kind":"found_in_text","path":"jaxrl/agents/sac/sac_learner.py","file_url":"https://github.com/proceduralia/high_replay_ratio_continuous_control/blob/HEAD/jaxrl/agents/sac/sac_learner.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":"09d56aa25dec1123"}},{"code_sha256_prefix":"ed8a594097bbbd78","entry":"_do_multiple_updates","repo":"proceduralia/high_replay_ratio_continuous_control","repo_kind":"found_in_text","path":"jaxrl/agents/sac/sac_learner.py","file_url":"https://github.com/proceduralia/high_replay_ratio_continuous_control/blob/HEAD/jaxrl/agents/sac/sac_learner.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":"ed8a594097bbbd78"}},{"code_sha256_prefix":"390709e4e4f378c9","entry":"_do_multiple_updates","repo":"naumix/BiggerRegularizedOptimistic","repo_kind":"official","path":"jaxrl/bro/bro_learner.py","file_url":"https://github.com/naumix/BiggerRegularizedOptimistic/blob/HEAD/jaxrl/bro/bro_learner.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":"390709e4e4f378c9"}},{"code_sha256_prefix":"1f40e01b920c066f","entry":"_update","repo":"proceduralia/high_replay_ratio_continuous_control","repo_kind":"found_in_text","path":"jaxrl/agents/sac/sac_learner.py","file_url":"https://github.com/proceduralia/high_replay_ratio_continuous_control/blob/HEAD/jaxrl/agents/sac/sac_learner.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":"1f40e01b920c066f"}},{"code_sha256_prefix":"9e582c32acd1ef30","entry":"_update","repo":"naumix/BiggerRegularizedOptimistic","repo_kind":"official","path":"jaxrl/bro/bro_learner.py","file_url":"https://github.com/naumix/BiggerRegularizedOptimistic/blob/HEAD/jaxrl/bro/bro_learner.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":"9e582c32acd1ef30"}},{"code_sha256_prefix":"ea6955fd0f62ad14","entry":"calculate_quantile_huber_loss","repo":"naumix/BiggerRegularizedOptimistic","repo_kind":"official","path":"jaxrl/bro/bro_learner.py","file_url":"https://github.com/naumix/BiggerRegularizedOptimistic/blob/HEAD/jaxrl/bro/bro_learner.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":"ea6955fd0f62ad14"}},{"code_sha256_prefix":"804f231f90e60d49","entry":"target_update","repo":"proceduralia/high_replay_ratio_continuous_control","repo_kind":"found_in_text","path":"jaxrl/agents/sac/sac_learner.py","file_url":"https://github.com/proceduralia/high_replay_ratio_continuous_control/blob/HEAD/jaxrl/agents/sac/sac_learner.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":"804f231f90e60d49"}},{"code_sha256_prefix":"ce51f9f65a9e0a89","entry":"target_update","repo":"naumix/BiggerRegularizedOptimistic","repo_kind":"official","path":"jaxrl/bro/bro_learner.py","file_url":"https://github.com/naumix/BiggerRegularizedOptimistic/blob/HEAD/jaxrl/bro/bro_learner.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":"ce51f9f65a9e0a89"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}