{"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/pre-training-with-synthetic-data-helps","title":"Pre-training with Synthetic Data Helps Offline Reinforcement Learning","arxiv_id":"2310.00771","date":"2023-10-01","proceeding":null,"authors":["Zecheng Wang","Che Wang","Zixuan Dong","Keith Ross"],"abstract":"Recently, it has been shown that for offline deep reinforcement learning (DRL), pre-training Decision Transformer with a large language corpus can improve downstream performance (Reid et al., 2022). A natural question to ask is whether this performance gain can only be achieved with language pre-training, or can be achieved with simpler pre-training schemes which do not involve language. In this paper, we first show that language is not essential for improved performance, and indeed pre-training with synthetic IID data for a small number of updates can match the performance gains from pre-training with a large language corpus; moreover, pre-training with data generated by a one-step Markov chain can further improve the performance. Inspired by these experimental results, we then consider pre-training Conservative Q-Learning (CQL), a popular offline DRL algorithm, which is Q-learning-based and typically employs a Multi-Layer Perceptron (MLP) backbone. Surprisingly, pre-training with simple synthetic data for a small number of updates can also improve CQL, providing consistent performance improvement on D4RL Gym locomotion datasets. The results of this paper not only illustrate the importance of pre-training for offline DRL but also show that the pre-training data can be synthetic and generated with remarkably simple mechanisms.","url_abs":"https://arxiv.org/abs/2310.00771v4","url_pdf":"https://arxiv.org/pdf/2310.00771v4.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":"pre-training-with-synthetic-data-helps","repo_url":"https://github.com/victor-wang-902/synthetic-pretrain-rl","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":0,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"d4rl","task_name":"D4RL"},{"task_slug":"deep-reinforcement-learning","task_name":"Deep Reinforcement Learning"},{"task_slug":"q-learning","task_name":"Q-Learning"},{"task_slug":"reinforcement-learning","task_name":"Reinforcement Learning"},{"task_slug":"reinforcement-learning-2","task_name":"reinforcement-learning"}],"methods":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"dense-connections","method_name":"Dense Connections"},{"method_slug":"dropout","method_name":"Dropout"},{"method_slug":"label-smoothing","method_name":"Label Smoothing"},{"method_slug":"layer-normalization","method_name":"Layer Normalization"},{"method_slug":"linear-layer","method_name":"Linear Layer"},{"method_slug":"multi-head-attention","method_name":"Multi-Head Attention"},{"method_slug":"position-wise-feed-forward-layer","method_name":"Position-Wise Feed-Forward Layer"},{"method_slug":"q-learning","method_name":"Q-Learning"},{"method_slug":"residual-connection","method_name":"Residual Connection"},{"method_slug":"softmax","method_name":"Softmax"},{"method_slug":"transformer","method_name":"Transformer"}],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"atlas_url":"https://app.syntology.ai/?focus=2310.00771","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2310.00771"}},"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/victor-wang-902/synthetic-pretrain-rl","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran":7},"by_repo_kind":{"official":{"samples":7,"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":0,"samples":[{"code_sha256_prefix":"7a02080f88f555ba","entry":"eval","repo":"victor-wang-902/synthetic-pretrain-rl","repo_kind":"official","path":"syn_pretraining/pretrain/pretrain_dist.py","file_url":"https://github.com/victor-wang-902/synthetic-pretrain-rl/blob/HEAD/syn_pretraining/pretrain/pretrain_dist.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":"7a02080f88f555ba"}},{"code_sha256_prefix":"75c95430710dc5f3","entry":"generate_mdp_data","repo":"victor-wang-902/synthetic-pretrain-rl","repo_kind":"official","path":"syn_pretraining_cql/pretrain_cql.py","file_url":"https://github.com/victor-wang-902/synthetic-pretrain-rl/blob/HEAD/syn_pretraining_cql/pretrain_cql.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":"75c95430710dc5f3"}},{"code_sha256_prefix":"82ccd092debb951c","entry":"get_dataloader","repo":"victor-wang-902/synthetic-pretrain-rl","repo_kind":"official","path":"syn_pretraining/pretrain/data.py","file_url":"https://github.com/victor-wang-902/synthetic-pretrain-rl/blob/HEAD/syn_pretraining/pretrain/data.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":"82ccd092debb951c"}},{"code_sha256_prefix":"c75faf5e36f4ca99","entry":"load_synthetic_dataset","repo":"victor-wang-902/synthetic-pretrain-rl","repo_kind":"official","path":"syn_pretraining/pretrain/data.py","file_url":"https://github.com/victor-wang-902/synthetic-pretrain-rl/blob/HEAD/syn_pretraining/pretrain/data.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":"c75faf5e36f4ca99"}},{"code_sha256_prefix":"b1521ca072cfd797","entry":"set_dt_args","repo":"victor-wang-902/synthetic-pretrain-rl","repo_kind":"official","path":"syn_pretraining/pretrain/pretrain_dist.py","file_url":"https://github.com/victor-wang-902/synthetic-pretrain-rl/blob/HEAD/syn_pretraining/pretrain/pretrain_dist.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":"b1521ca072cfd797"}},{"code_sha256_prefix":"7ae90f4ad90c9da4","entry":"softmax_with_torch","repo":"victor-wang-902/synthetic-pretrain-rl","repo_kind":"official","path":"syn_pretraining_cql/pretrain_cql.py","file_url":"https://github.com/victor-wang-902/synthetic-pretrain-rl/blob/HEAD/syn_pretraining_cql/pretrain_cql.py","link_basis":"first_harvest_node","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":"7ae90f4ad90c9da4"}},{"code_sha256_prefix":"e4035f8443c3f9ee","entry":"subsample_batch","repo":"victor-wang-902/synthetic-pretrain-rl","repo_kind":"official","path":"syn_pretraining_cql/pretrain_cql.py","file_url":"https://github.com/victor-wang-902/synthetic-pretrain-rl/blob/HEAD/syn_pretraining_cql/pretrain_cql.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":"e4035f8443c3f9ee"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}