{"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/amago-2-breaking-the-multi-task-barrier-in","title":"AMAGO-2: Breaking the Multi-Task Barrier in Meta-Reinforcement Learning with Transformers","arxiv_id":"2411.11188","date":"2024-11-17","proceeding":null,"authors":["Jake Grigsby","Justin Sasek","Samyak Parajuli","Daniel Adebi","Amy Zhang","Yuke Zhu"],"abstract":"Language models trained on diverse datasets unlock generalization by in-context learning. Reinforcement Learning (RL) policies can achieve a similar effect by meta-learning within the memory of a sequence model. However, meta-RL research primarily focuses on adapting to minor variations of a single task. It is difficult to scale towards more general behavior without confronting challenges in multi-task optimization, and few solutions are compatible with meta-RL's goal of learning from large training sets of unlabeled tasks. To address this challenge, we revisit the idea that multi-task RL is bottlenecked by imbalanced training losses created by uneven return scales across different tasks. We build upon recent advancements in Transformer-based (in-context) meta-RL and evaluate a simple yet scalable solution where both an agent's actor and critic objectives are converted to classification terms that decouple optimization from the current scale of returns. Large-scale comparisons in Meta-World ML45, Multi-Game Procgen, Multi-Task POPGym, Multi-Game Atari, and BabyAI find that this design unlocks significant progress in online multi-task adaptation and memory problems without explicit task labels.","url_abs":"https://arxiv.org/abs/2411.11188v1","url_pdf":"https://arxiv.org/pdf/2411.11188v1.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":"amago-2-breaking-the-multi-task-barrier-in","repo_url":"https://github.com/ut-austin-rpl/amago","is_official":1,"mentioned_in_paper":0,"mentioned_in_github":1,"framework":"pytorch","reach":{"status":"ok","spdx":"MIT"}}],"tasks":[{"task_slug":"in-context-learning","task_name":"In-Context Learning"},{"task_slug":"meta-reinforcement-learning","task_name":"Meta Reinforcement Learning"},{"task_slug":"meta-learning","task_name":"Meta-Learning"},{"task_slug":"reinforcement-learning-1","task_name":"Reinforcement Learning (RL)"}],"methods":[],"datasets_introduced":[],"methods_introduced":[],"results":[],"syntology":{"syntology_url":"https://syntology.ai/paper/2411.11188","atlas_url":"https://app.syntology.ai/?focus=2411.11188","mcp":{"get_harvested_code_for_paper":{"arxiv_id":"2411.11188"}},"developers":"https://syntology.ai/developers","read_at":"2026-09-25T09:33:49+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/ut-austin-rpl/amago","reach":{"status":"ok","spdx":"MIT"}}],"summary":{"ran_draft_wrong":2,"unverified":8},"by_repo_kind":{"official":{"samples":9,"ran":1,"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":1,"samples":[{"code_sha256_prefix":"0813fc39f3d9ebb0","entry":"add_cli","repo":null,"repo_kind":null,"path":null,"file_url":null,"link_basis":"identical_code_first_harvested_elsewhere","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":"deterministic","behaviour_fingerprint":false,"licence":null,"inline_ok":false,"mcp_get_code":{"code_sha256":"0813fc39f3d9ebb0"}},{"code_sha256_prefix":"cb28542e368fc51b","entry":"register_agent","repo":"ut-austin-rpl/amago","repo_kind":"official","path":"amago/agent.py","file_url":"https://github.com/ut-austin-rpl/amago/blob/HEAD/amago/agent.py","link_basis":"plan_row","language":"python","status":"ran_draft_wrong","verification_level":1,"contract_check":"OUTPUT_MISDECLARED","metamorphic_tier":null,"behaviour_fingerprint":false,"licence":"MIT","inline_ok":true,"mcp_get_code":{"code_sha256":"cb28542e368fc51b"}},{"code_sha256_prefix":"9069125b4b210054","entry":"RLData_pad_collate","repo":"ut-austin-rpl/amago","repo_kind":"official","path":"amago/loading.py","file_url":"https://github.com/ut-austin-rpl/amago/blob/HEAD/amago/loading.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":"9069125b4b210054"}},{"code_sha256_prefix":"2ffee777eb103fb3","entry":"get_action_dim_and_type","repo":"ut-austin-rpl/amago","repo_kind":"official","path":"amago/agent.py","file_url":"https://github.com/ut-austin-rpl/amago/blob/HEAD/amago/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":"2ffee777eb103fb3"}},{"code_sha256_prefix":"ae3524026f11fa9f","entry":"get_agent_cls","repo":"ut-austin-rpl/amago","repo_kind":"official","path":"amago/agent.py","file_url":"https://github.com/ut-austin-rpl/amago/blob/HEAD/amago/agent.py","link_basis":"plan_row","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":"ae3524026f11fa9f"}},{"code_sha256_prefix":"7176a351fc668593","entry":"get_path_to_trajs","repo":"ut-austin-rpl/amago","repo_kind":"official","path":"amago/loading.py","file_url":"https://github.com/ut-austin-rpl/amago/blob/HEAD/amago/loading.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":"7176a351fc668593"}},{"code_sha256_prefix":"0f3f64d4db295213","entry":"get_traj_encoder_cls","repo":"ut-austin-rpl/amago","repo_kind":"official","path":"amago/nets/traj_encoders.py","file_url":"https://github.com/ut-austin-rpl/amago/blob/HEAD/amago/nets/traj_encoders.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":"0f3f64d4db295213"}},{"code_sha256_prefix":"60a00fe723fb7874","entry":"register_traj_encoder","repo":"ut-austin-rpl/amago","repo_kind":"official","path":"amago/nets/traj_encoders.py","file_url":"https://github.com/ut-austin-rpl/amago/blob/HEAD/amago/nets/traj_encoders.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":"60a00fe723fb7874"}},{"code_sha256_prefix":"9f588318c9339dd5","entry":"softplus_bounded_positive","repo":"ut-austin-rpl/amago","repo_kind":"official","path":"amago/nets/policy_dists.py","file_url":"https://github.com/ut-austin-rpl/amago/blob/HEAD/amago/nets/policy_dists.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":"9f588318c9339dd5"}},{"code_sha256_prefix":"679514cf7e0c892c","entry":"tanh_bounded_positive","repo":"ut-austin-rpl/amago","repo_kind":"official","path":"amago/nets/policy_dists.py","file_url":"https://github.com/ut-austin-rpl/amago/blob/HEAD/amago/nets/policy_dists.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":"679514cf7e0c892c"}}]},"arxiv_metadata":null,"syntology_extracted_results":null}