{"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/the-curse-of-diversity-in-ensemble-based","title":"The Curse of Diversity in Ensemble-Based Exploration","arxiv_id":"2405.04342","date":"2024-05-07","proceeding":null,"authors":["Zhixuan Lin","Pierluca D'Oro","Evgenii Nikishin","Aaron Courville"],"abstract":"We uncover a surprising phenomenon in deep reinforcement learning: training a diverse ensemble of data-sharing agents -- a well-established exploration strategy -- can significantly impair the performance of the individual ensemble members when compared to standard single-agent training. Through careful analysis, we attribute the degradation in performance to the low proportion of self-generated data in the shared training data for each ensemble member, as well as the inefficiency of the individual ensemble members to learn from such highly off-policy data. We thus name this phenomenon the curse of diversity. We find that several intuitive solutions -- such as a larger replay buffer or a smaller ensemble size -- either fail to consistently mitigate the performance loss or undermine the advantages of ensembling. Finally, we demonstrate the potential of representation learning to counteract the curse of diversity with a novel method named Cross-Ensemble Representation Learning (CERL) in both discrete and continuous control domains. Our work offers valuable insights into an unexpected pitfall in ensemble-based exploration and raises important caveats for future applications of similar approaches.","url_abs":"https://arxiv.org/abs/2405.04342v1","url_pdf":"https://arxiv.org/pdf/2405.04342v1.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":"the-curse-of-diversity-in-ensemble-based","repo_url":"https://github.com/zhixuan-lin/ensemble-rl-continuous","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"jax","reach":null},{"paper_slug":"the-curse-of-diversity-in-ensemble-based","repo_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"jax","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"attribute","task_name":"Attribute"},{"task_slug":"continuous-control","task_name":"Continuous Control"},{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"diversity","task_name":"Diversity"},{"task_slug":"representation-learning","task_name":"Representation Learning"},{"task_slug":"continuous-control","task_name":"continuous-control"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2405.04342","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2405.04342"}},"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/zhixuan-lin/ensemble-rl-discrete","reach":{"status":"ok","spdx":"MIT"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/zhixuan-lin/ensemble-rl-continuous","reach":null}],"summary":{"ran":2,"ran_honours":1,"unverified":13},"by_repo_kind":{"official":{"samples":16,"ran":3,"repositories":2}},"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":"ee6d2e618cac54d1","entry":"get_gammas","repo":"zhixuan-lin/ensemble-rl-discrete","repo_kind":"official","path":"ensemble_rl/agents/boot_dqn_agent.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete/blob/HEAD/ensemble_rl/agents/boot_dqn_agent.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":"ee6d2e618cac54d1"}},{"code_sha256_prefix":"2ec107672ef4af91","entry":"get_param_count","repo":"zhixuan-lin/ensemble-rl-continuous","repo_kind":"official","path":"jaxrl/agents/cel/cel_learner.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-continuous/blob/HEAD/jaxrl/agents/cel/cel_learner.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"2ec107672ef4af91"}},{"code_sha256_prefix":"66a8c225827dbcfc","entry":"squarify","repo":"zhixuan-lin/ensemble-rl-discrete","repo_kind":"official","path":"ensemble_rl/metric_utils.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete/blob/HEAD/ensemble_rl/metric_utils.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":"66a8c225827dbcfc"}},{"code_sha256_prefix":"373f13eb372c6f1c","entry":"CELLearner","repo":"zhixuan-lin/ensemble-rl-continuous","repo_kind":"official","path":"jaxrl/agents/cel/cel_learner.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-continuous/blob/HEAD/jaxrl/agents/cel/cel_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":"373f13eb372c6f1c"}},{"code_sha256_prefix":"a681d1d465ca2192","entry":"Model","repo":"zhixuan-lin/ensemble-rl-continuous","repo_kind":"official","path":"jaxrl/agents/cel/cel_learner.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-continuous/blob/HEAD/jaxrl/agents/cel/cel_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":"a681d1d465ca2192"}},{"code_sha256_prefix":"8078c02fd29a891c","entry":"_sample_actions","repo":"zhixuan-lin/ensemble-rl-continuous","repo_kind":"official","path":"jaxrl/agents/cel/cel_learner.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-continuous/blob/HEAD/jaxrl/agents/cel/cel_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":"8078c02fd29a891c"}},{"code_sha256_prefix":"e4e6020a9a6a6bbb","entry":"_sample_index","repo":"zhixuan-lin/ensemble-rl-continuous","repo_kind":"official","path":"jaxrl/agents/cel/cel_learner.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-continuous/blob/HEAD/jaxrl/agents/cel/cel_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":"e4e6020a9a6a6bbb"}},{"code_sha256_prefix":"61fd4828e83d0558","entry":"_update_jit","repo":"zhixuan-lin/ensemble-rl-continuous","repo_kind":"official","path":"jaxrl/agents/cel/cel_learner.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-continuous/blob/HEAD/jaxrl/agents/cel/cel_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":"61fd4828e83d0558"}},{"code_sha256_prefix":"389e520afe20c705","entry":"cosine_distance","repo":"zhixuan-lin/ensemble-rl-discrete","repo_kind":"official","path":"ensemble_rl/metric_utils.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete/blob/HEAD/ensemble_rl/metric_utils.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":"389e520afe20c705"}},{"code_sha256_prefix":"9ebd5c538050adfc","entry":"l2","repo":"zhixuan-lin/ensemble-rl-discrete","repo_kind":"official","path":"ensemble_rl/metric_utils.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete/blob/HEAD/ensemble_rl/metric_utils.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":"9ebd5c538050adfc"}},{"code_sha256_prefix":"5eff8223ef2a7296","entry":"select_action","repo":"zhixuan-lin/ensemble-rl-discrete","repo_kind":"official","path":"ensemble_rl/agents/boot_quantile_agent.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete/blob/HEAD/ensemble_rl/agents/boot_quantile_agent.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":"5eff8223ef2a7296"}},{"code_sha256_prefix":"4e8df17f4dcd90d6","entry":"target_distribution","repo":"zhixuan-lin/ensemble-rl-discrete","repo_kind":"official","path":"ensemble_rl/agents/boot_quantile_agent.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete/blob/HEAD/ensemble_rl/agents/boot_quantile_agent.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":"4e8df17f4dcd90d6"}},{"code_sha256_prefix":"9697728fcd0cfeb3","entry":"target_q","repo":"zhixuan-lin/ensemble-rl-discrete","repo_kind":"official","path":"ensemble_rl/agents/boot_dqn_agent.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete/blob/HEAD/ensemble_rl/agents/boot_dqn_agent.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":"9697728fcd0cfeb3"}},{"code_sha256_prefix":"336b5ba1af3f1cd7","entry":"target_q","repo":"zhixuan-lin/ensemble-rl-discrete","repo_kind":"official","path":"ensemble_rl/agents/dqn_agent.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete/blob/HEAD/ensemble_rl/agents/dqn_agent.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":"336b5ba1af3f1cd7"}},{"code_sha256_prefix":"d7b5436efe6a474c","entry":"target_update","repo":"zhixuan-lin/ensemble-rl-continuous","repo_kind":"official","path":"jaxrl/agents/cel/cel_learner.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-continuous/blob/HEAD/jaxrl/agents/cel/cel_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":"d7b5436efe6a474c"}},{"code_sha256_prefix":"fed787981f691a19","entry":"train","repo":"zhixuan-lin/ensemble-rl-discrete","repo_kind":"official","path":"ensemble_rl/agents/boot_quantile_agent.py","file_url":"https://github.com/zhixuan-lin/ensemble-rl-discrete/blob/HEAD/ensemble_rl/agents/boot_quantile_agent.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":"fed787981f691a19"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}