{"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/transformers-are-sample-efficient-world","title":"Transformers are Sample-Efficient World Models","arxiv_id":"2209.00588","date":"2022-09-01","proceeding":null,"authors":["Vincent Micheli","Eloi Alonso","François Fleuret"],"abstract":"Deep reinforcement learning agents are notoriously sample inefficient, which considerably limits their application to real-world problems. Recently, many model-based methods have been designed to address this issue, with learning in the imagination of a world model being one of the most prominent approaches. However, while virtually unlimited interaction with a simulated environment sounds appealing, the world model has to be accurate over extended periods of time. Motivated by the success of Transformers in sequence modeling tasks, we introduce IRIS, a data-efficient agent that learns in a world model composed of a discrete autoencoder and an autoregressive Transformer. With the equivalent of only two hours of gameplay in the Atari 100k benchmark, IRIS achieves a mean human normalized score of 1.046, and outperforms humans on 10 out of 26 games, setting a new state of the art for methods without lookahead search. To foster future research on Transformers and world models for sample-efficient reinforcement learning, we release our code and models at https://github.com/eloialonso/iris.","url_abs":"https://arxiv.org/abs/2209.00588v2","url_pdf":"https://arxiv.org/pdf/2209.00588v2.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":"transformers-are-sample-efficient-world","repo_url":"https://github.com/eloialonso/iris","is_official":1,"mentioned_in_paper":1,"mentioned_in_github":1,"framework":"pytorch","reach":null},{"paper_slug":"transformers-are-sample-efficient-world","repo_url":"https://github.com/vmicheli/delta-iris","is_official":0,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"GPL-3.0"}}],"tasks":[{"task_slug":"atari-games-100k","task_name":"Atari Games 100k"},{"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":[{"method_slug":"absolute-position-encodings","method_name":"Absolute Position Encodings"},{"method_slug":"adam","method_name":"Adam"},{"method_slug":"attention","method_name":"Attention"},{"method_slug":"average-pooling","method_name":"Average Pooling"},{"method_slug":"bpe","method_name":"BPE"},{"method_slug":"batch-normalization","method_name":"Batch Normalization"},{"method_slug":"convolution","method_name":"Convolution"},{"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":"monte-carlo-tree-search","method_name":"Monte-Carlo Tree Search"},{"method_slug":"muzero","method_name":"MuZero"},{"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":"prioritized-experience-replay","method_name":"Prioritized Experience Replay"},{"method_slug":"relu","method_name":"ReLU"},{"method_slug":"residual-block","method_name":"Residual Block"},{"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=2209.00588","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2209.00588"}},"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/eloialonso/iris","reach":null},{"provenance":"external:paperswithcode_snapshot_2025-07-28","url":"https://github.com/vmicheli/delta-iris","reach":{"status":"ok","spdx":"GPL-3.0"}}],"summary":{"ran":15,"ran_draft_wrong":1,"ran_honours":1,"unverified":9},"by_repo_kind":{"official":{"samples":26,"ran":17,"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":26,"samples":[{"code_sha256_prefix":"7f939e4da5014501","entry":"AssignWithoutInplaceCheck","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"7f939e4da5014501"}},{"code_sha256_prefix":"9d1f4ad302b969e4","entry":"AttnBlock","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"9d1f4ad302b969e4"}},{"code_sha256_prefix":"256e6e8e5c0039da","entry":"Cache","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"256e6e8e5c0039da"}},{"code_sha256_prefix":"e8df5d8fcd9e9339","entry":"Downsample","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":true,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"e8df5d8fcd9e9339"}},{"code_sha256_prefix":"2def3130991aaf3c","entry":"Embedder","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"2def3130991aaf3c"}},{"code_sha256_prefix":"cd65aa1097b08332","entry":"EncoderDecoderConfig","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"cd65aa1097b08332"}},{"code_sha256_prefix":"b4868fbed7dce84a","entry":"Head","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"b4868fbed7dce84a"}},{"code_sha256_prefix":"93bd9c4b1f02db59","entry":"KVCache","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"93bd9c4b1f02db59"}},{"code_sha256_prefix":"fba3ef4043569d6f","entry":"KeysValues","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"fba3ef4043569d6f"}},{"code_sha256_prefix":"d7a1b7c2fadc881a","entry":"LossWithIntermediateLosses","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"d7a1b7c2fadc881a"}},{"code_sha256_prefix":"41ffde6c773af0d6","entry":"Normalize","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.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":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"41ffde6c773af0d6"}},{"code_sha256_prefix":"3e962c6d7a040002","entry":"Slicer","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"3e962c6d7a040002"}},{"code_sha256_prefix":"bf4dc4c0382a6bd0","entry":"TokenizerEncoderOutput","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"bf4dc4c0382a6bd0"}},{"code_sha256_prefix":"06c7be0a3042efdc","entry":"TransformerConfig","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"06c7be0a3042efdc"}},{"code_sha256_prefix":"53918c7471cbccdc","entry":"Upsample","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":true,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"53918c7471cbccdc"}},{"code_sha256_prefix":"419935433daf0f1b","entry":"WorldModelOutput","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran","verification_level":1,"contract_check":null,"metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"419935433daf0f1b"}},{"code_sha256_prefix":"3f6efbafceb71819","entry":"nonlinearity","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"ran_honours","verification_level":1,"contract_check":"HONOURS","metamorphic_tier":"invariant","behaviour_fingerprint":true,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"3f6efbafceb71819"}},{"code_sha256_prefix":"ee23a284027c25be","entry":"Block","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"ee23a284027c25be"}},{"code_sha256_prefix":"4e93d0d0e4839243","entry":"Decoder","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"4e93d0d0e4839243"}},{"code_sha256_prefix":"aee0695eaeb766c4","entry":"Encoder","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"aee0695eaeb766c4"}},{"code_sha256_prefix":"caffc15eb259ebb0","entry":"ResnetBlock","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"caffc15eb259ebb0"}},{"code_sha256_prefix":"f16fd90513419812","entry":"SelfAttention","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"f16fd90513419812"}},{"code_sha256_prefix":"79b0b0b918ab2ac4","entry":"Tokenizer","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"79b0b0b918ab2ac4"}},{"code_sha256_prefix":"d65bf756384c37cd","entry":"Transformer","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"d65bf756384c37cd"}},{"code_sha256_prefix":"cfb3154cdccd72da","entry":"WorldModel","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"cfb3154cdccd72da"}},{"code_sha256_prefix":"7fc674de9a0a0a02","entry":"init_weights","repo":"eloialonso/iris","repo_kind":"official","path":"src/models/world_model.py","file_url":"https://github.com/eloialonso/iris/blob/HEAD/src/models/world_model.py","link_basis":"first_harvest_node","language":"python","status":"unverified","verification_level":0,"contract_check":null,"metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"GPL-3.0","inline_ok":false,"mcp_get_code":{"code_sha256":"7fc674de9a0a0a02"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}