{"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/on-pathologies-in-kl-regularized-1","title":"On Pathologies in KL-Regularized Reinforcement Learning from Expert Demonstrations","arxiv_id":"2212.13936","date":"2022-12-28","proceeding":"NeurIPS 2021 12","authors":["Tim G. J. Rudner","Cong Lu","Michael A. Osborne","Yarin Gal","Yee Whye Teh"],"abstract":"KL-regularized reinforcement learning from expert demonstrations has proved successful in improving the sample efficiency of deep reinforcement learning algorithms, allowing them to be applied to challenging physical real-world tasks. However, we show that KL-regularized reinforcement learning with behavioral reference policies derived from expert demonstrations can suffer from pathological training dynamics that can lead to slow, unstable, and suboptimal online learning. We show empirically that the pathology occurs for commonly chosen behavioral policy classes and demonstrate its impact on sample efficiency and online policy performance. Finally, we show that the pathology can be remedied by non-parametric behavioral reference policies and that this allows KL-regularized reinforcement learning to significantly outperform state-of-the-art approaches on a variety of challenging locomotion and dexterous hand manipulation tasks.","url_abs":"https://arxiv.org/abs/2212.13936v1","url_pdf":"https://arxiv.org/pdf/2212.13936v1.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":"on-pathologies-in-kl-regularized-1","repo_url":"https://github.com/conglu1997/nppac","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2212.13936","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2212.13936"}},"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":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/conglu1997/nppac","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":5,"ran_honours":1,"unverified":1},"by_repo_kind":{"official":{"samples":7,"ran":6,"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":"93d28ca67303e04b","entry":"clip_magnitude","repo":"conglu1997/nppac","repo_kind":"official","path":"rlkit/pythonplusplus.py","file_url":"https://github.com/conglu1997/nppac/blob/HEAD/rlkit/pythonplusplus.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"93d28ca67303e04b"}},{"code_sha256_prefix":"fadf5f762ff04cd6","entry":"create_stats_ordered_dict","repo":"conglu1997/nppac","repo_kind":"official","path":"rlkit/core/eval_util.py","file_url":"https://github.com/conglu1997/nppac/blob/HEAD/rlkit/core/eval_util.py","link_basis":"harvester_set","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":"fadf5f762ff04cd6"}},{"code_sha256_prefix":"a5d10a0b704c8db7","entry":"get_average_returns","repo":"conglu1997/nppac","repo_kind":"official","path":"rlkit/core/eval_util.py","file_url":"https://github.com/conglu1997/nppac/blob/HEAD/rlkit/core/eval_util.py","link_basis":"harvester_set","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":"a5d10a0b704c8db7"}},{"code_sha256_prefix":"9910e2fc297f8665","entry":"identity","repo":"conglu1997/nppac","repo_kind":"official","path":"rlkit/pythonplusplus.py","file_url":"https://github.com/conglu1997/nppac/blob/HEAD/rlkit/pythonplusplus.py","link_basis":"harvester_set","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"9910e2fc297f8665"}},{"code_sha256_prefix":"4138766e854db009","entry":"is_numeric","repo":"conglu1997/nppac","repo_kind":"official","path":"rlkit/pythonplusplus.py","file_url":"https://github.com/conglu1997/nppac/blob/HEAD/rlkit/pythonplusplus.py","link_basis":"harvester_set","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"4138766e854db009"}},{"code_sha256_prefix":"5cb0c55f58279f59","entry":"simple_separated_format","repo":"conglu1997/nppac","repo_kind":"official","path":"rlkit/core/tabulate.py","file_url":"https://github.com/conglu1997/nppac/blob/HEAD/rlkit/core/tabulate.py","link_basis":"harvester_set","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":"5cb0c55f58279f59"}},{"code_sha256_prefix":"6537ea2eace191a6","entry":"batch_assess","repo":"conglu1997/nppac","repo_kind":"official","path":"nppac/clone_from_dataset_gp.py","file_url":"https://github.com/conglu1997/nppac/blob/HEAD/nppac/clone_from_dataset_gp.py","link_basis":"harvester_set","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":"6537ea2eace191a6"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}