{"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/conservative-objective-models-for-effective","title":"Conservative Objective Models for Effective Offline Model-Based Optimization","arxiv_id":"2107.06882","date":"2021-07-14","proceeding":null,"authors":["Brandon Trabucco","Aviral Kumar","Xinyang Geng","Sergey Levine"],"abstract":"Computational design problems arise in a number of settings, from synthetic biology to computer architectures. In this paper, we aim to solve data-driven model-based optimization (MBO) problems, where the goal is to find a design input that maximizes an unknown objective function provided access to only a static dataset of prior experiments. Such data-driven optimization procedures are the only practical methods in many real-world domains where active data collection is expensive (e.g., when optimizing over proteins) or dangerous (e.g., when optimizing over aircraft designs). Typical methods for MBO that optimize the design against a learned model suffer from distributional shift: it is easy to find a design that \"fools\" the model into predicting a high value. To overcome this, we propose conservative objective models (COMs), a method that learns a model of the objective function that lower bounds the actual value of the ground-truth objective on out-of-distribution inputs, and uses it for optimization. Structurally, COMs resemble adversarial training methods used to overcome adversarial examples. COMs are simple to implement and outperform a number of existing methods on a wide range of MBO problems, including optimizing protein sequences, robot morphologies, neural network weights, and superconducting materials.","url_abs":"https://arxiv.org/abs/2107.06882v1","url_pdf":"https://arxiv.org/pdf/2107.06882v1.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":"conservative-objective-models-for-effective","repo_url":"https://github.com/brandontrabucco/design-baselines","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"tf","reach":null},{"paper_slug":"conservative-objective-models-for-effective","repo_url":"https://github.com/rail-berkeley/design-baselines","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"tf","reach":null}],"tasks":[],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2107.06882","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2107.06882"}},"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/brandontrabucco/design-baselines","reach":null},{"provenance":"deterministic:regex_extraction","url":"https://github.com/dhbrookes/CbAS","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/rail-berkeley/design-baselines","reach":null}],"summary":{"ran":2,"ran_draft_wrong":3,"ran_honours":2,"ran_violates":1,"unverified":5},"by_repo_kind":{"listed":{"samples":3,"ran":1,"repositories":1},"found_in_text":{"samples":10,"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":10,"samples":[{"code_sha256_prefix":"fabe5296d2a0d197","entry":"SimpleVAE","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.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":"fabe5296d2a0d197"}},{"code_sha256_prefix":"29e1a3bca6e404e7","entry":"color","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.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":"29e1a3bca6e404e7"}},{"code_sha256_prefix":"a80593f2798a9126","entry":"convert_idx_array_to_aas","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"a80593f2798a9126"}},{"code_sha256_prefix":"7615a7c2a742b9c4","entry":"get_balaji_predictions","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"7615a7c2a742b9c4"}},{"code_sha256_prefix":"95e9b9ea7dfff908","entry":"get_rank","repo":"rail-berkeley/design-baselines","repo_kind":"listed","path":"design_baselines/coms_cleaned/trainers.py","file_url":"https://github.com/rail-berkeley/design-baselines/blob/HEAD/design_baselines/coms_cleaned/trainers.py","link_basis":"first_harvest_node","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":"95e9b9ea7dfff908"}},{"code_sha256_prefix":"48906955911d7219","entry":"get_samples","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.py","link_basis":"first_harvest_node","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"48906955911d7219"}},{"code_sha256_prefix":"ac92d5e9383fe2b9","entry":"identity_loss","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"ac92d5e9383fe2b9"}},{"code_sha256_prefix":"c67022b5a8beb86a","entry":"summed_categorical_crossentropy","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.py","link_basis":"first_harvest_node","language":"python","status":"ran_violates","verification_level":1,"contract_check":"VIOLATES","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"NONE","inline_ok":false,"mcp_get_code":{"code_sha256":"c67022b5a8beb86a"}},{"code_sha256_prefix":"2ce5de7e7bad4706","entry":"BaseVAE","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.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":"2ce5de7e7bad4706"}},{"code_sha256_prefix":"1bf1af505277c864","entry":"ConservativeObjectiveModel","repo":"rail-berkeley/design-baselines","repo_kind":"listed","path":"design_baselines/coms_cleaned/trainers.py","file_url":"https://github.com/rail-berkeley/design-baselines/blob/HEAD/design_baselines/coms_cleaned/trainers.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":"1bf1af505277c864"}},{"code_sha256_prefix":"ecb6882617cc76da","entry":"build_vae","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.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":"ecb6882617cc76da"}},{"code_sha256_prefix":"89071da4c8ba9b55","entry":"spearman","repo":"rail-berkeley/design-baselines","repo_kind":"listed","path":"design_baselines/coms_cleaned/trainers.py","file_url":"https://github.com/rail-berkeley/design-baselines/blob/HEAD/design_baselines/coms_cleaned/trainers.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":"89071da4c8ba9b55"}},{"code_sha256_prefix":"d12e787e592de8d7","entry":"weighted_ml_opt","repo":"dhbrookes/CbAS","repo_kind":"found_in_text","path":"src/optimization_algs.py","file_url":"https://github.com/dhbrookes/CbAS/blob/HEAD/src/optimization_algs.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":"d12e787e592de8d7"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}