{"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/strictly-batch-imitation-learning-by-energy","title":"Strictly Batch Imitation Learning by Energy-based Distribution Matching","arxiv_id":"2006.14154","date":"2020-06-25","proceeding":"NeurIPS 2020 12","authors":["Daniel Jarrett","Ioana Bica","Mihaela van der Schaar"],"abstract":"Consider learning a policy purely on the basis of demonstrated behavior -- that is, with no access to reinforcement signals, no knowledge of transition dynamics, and no further interaction with the environment. This *strictly batch imitation learning* problem arises wherever live experimentation is costly, such as in healthcare. One solution is simply to retrofit existing algorithms for apprenticeship learning to work in the offline setting. But such an approach leans heavily on off-policy evaluation or offline model estimation, and can be indirect and inefficient. We argue that a good solution should be able to explicitly parameterize a policy (i.e. respecting action conditionals), implicitly learn from rollout dynamics (i.e. leveraging state marginals), and -- crucially -- operate in an entirely offline fashion. To address this challenge, we propose a novel technique by *energy-based distribution matching* (EDM): By identifying parameterizations of the (discriminative) model of a policy with the (generative) energy function for state distributions, EDM yields a simple but effective solution that equivalently minimizes a divergence between the occupancy measure for the demonstrator and a model thereof for the imitator. Through experiments with application to control and healthcare settings, we illustrate consistent performance gains over existing algorithms for strictly batch imitation learning.","url_abs":"https://arxiv.org/abs/2006.14154v2","url_pdf":"https://arxiv.org/pdf/2006.14154v2.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":"strictly-batch-imitation-learning-by-energy","repo_url":"https://github.com/vanderschaarlab/mlforhealthlabpub/tree/main/alg/edm","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":0,"framework":"jax","reach":null}],"tasks":[{"task_slug":"imitation-learning","task_name":"Imitation Learning"},{"task_slug":"off-policy-evaluation","task_name":"Off-policy evaluation"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2006.14154","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2006.14154"}},"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/google-research/google-research","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/danieljarrett/EDM","reach":{"status":"gone","observed_at":"2026-09-17","how":"tree_404+repo_404"}},{"provenance":"deterministic:regex_extraction","url":"https://github.com/wgrathwohl/JEM","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/dtak/batch-apprenticeship-learning","reach":{"status":"ok","spdx":"Apache-2.0"}},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/vanderschaarlab/mlforhealthlabpub/tree/main/alg/edm","reach":null}],"summary":{"ran":6,"ran_draft_wrong":2,"unverified":9},"by_repo_kind":{"found_in_text":{"samples":17,"ran":8,"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":"c22f2d08409192a0","entry":"ActNorm","repo":"wgrathwohl/JEM","repo_kind":"found_in_text","path":"train_wrn_ebm.py","file_url":"https://github.com/wgrathwohl/JEM/blob/HEAD/train_wrn_ebm.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c22f2d08409192a0"}},{"code_sha256_prefix":"4985cbdb3ad39e8c","entry":"F","repo":"wgrathwohl/JEM","repo_kind":"found_in_text","path":"train_wrn_ebm.py","file_url":"https://github.com/wgrathwohl/JEM/blob/HEAD/train_wrn_ebm.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"4985cbdb3ad39e8c"}},{"code_sha256_prefix":"fdbb7a963e661fb1","entry":"Identity","repo":"wgrathwohl/JEM","repo_kind":"found_in_text","path":"train_wrn_ebm.py","file_url":"https://github.com/wgrathwohl/JEM/blob/HEAD/train_wrn_ebm.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"fdbb7a963e661fb1"}},{"code_sha256_prefix":"84de817b353f7c2c","entry":"Wide_ResNet","repo":"wgrathwohl/JEM","repo_kind":"found_in_text","path":"train_wrn_ebm.py","file_url":"https://github.com/wgrathwohl/JEM/blob/HEAD/train_wrn_ebm.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"84de817b353f7c2c"}},{"code_sha256_prefix":"bf19b9ba510a9e4a","entry":"conv3x3","repo":"wgrathwohl/JEM","repo_kind":"found_in_text","path":"train_wrn_ebm.py","file_url":"https://github.com/wgrathwohl/JEM/blob/HEAD/train_wrn_ebm.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"bf19b9ba510a9e4a"}},{"code_sha256_prefix":"54752d9ecb18a6fb","entry":"get_norm","repo":"wgrathwohl/JEM","repo_kind":"found_in_text","path":"train_wrn_ebm.py","file_url":"https://github.com/wgrathwohl/JEM/blob/HEAD/train_wrn_ebm.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"54752d9ecb18a6fb"}},{"code_sha256_prefix":"aad5113d0b095522","entry":"shape_el","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"src/core/distributions.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/src/core/distributions.py","link_basis":"plan_row","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"aad5113d0b095522"}},{"code_sha256_prefix":"8ecb413d596ba546","entry":"wide_basic","repo":"wgrathwohl/JEM","repo_kind":"found_in_text","path":"train_wrn_ebm.py","file_url":"https://github.com/wgrathwohl/JEM/blob/HEAD/train_wrn_ebm.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"8ecb413d596ba546"}},{"code_sha256_prefix":"7e084b093f24aee5","entry":"f_eq","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"src/contrib/scirl.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/src/contrib/scirl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7e084b093f24aee5"}},{"code_sha256_prefix":"7613cdbabed93897","entry":"f_geq","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"src/contrib/scirl.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/src/contrib/scirl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"7613cdbabed93897"}},{"code_sha256_prefix":"01397f68828e5085","entry":"get_linear_basis","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"src/core/basis.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/src/core/basis.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"01397f68828e5085"}},{"code_sha256_prefix":"2fdd9847cb765877","entry":"get_next_observation","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"main_sepsis.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/main_sepsis.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"2fdd9847cb765877"}},{"code_sha256_prefix":"e86b3e45a52de298","entry":"make_pdtype","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"src/core/distributions.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/src/core/distributions.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"e86b3e45a52de298"}},{"code_sha256_prefix":"72780861907b8bdd","entry":"new_reset","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"src/core/gym_env.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/src/core/gym_env.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"72780861907b8bdd"}},{"code_sha256_prefix":"634baa09ecda13c4","entry":"new_timelimit_reset","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"src/core/gym_env.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/src/core/gym_env.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"634baa09ecda13c4"}},{"code_sha256_prefix":"59357864346147ca","entry":"non_scalar_vectorize","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"src/contrib/scirl.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/src/contrib/scirl.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"59357864346147ca"}},{"code_sha256_prefix":"c415e5b34587ae0f","entry":"stack_observation","repo":"dtak/batch-apprenticeship-learning","repo_kind":"found_in_text","path":"main_sepsis.py","file_url":"https://github.com/dtak/batch-apprenticeship-learning/blob/HEAD/main_sepsis.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"Apache-2.0","inline_ok":true,"mcp_get_code":{"code_sha256":"c415e5b34587ae0f"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}