{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:KU4DF3GNU7ZFWFCM2XGVF5XRBL","short_pith_number":"pith:KU4DF3GN","schema_version":"1.0","canonical_sha256":"553832eccda7f25b144cd5cd52f6f10ac8bcb91531144a925a75faea6fe46a94","source":{"kind":"arxiv","id":"2509.11233","version":1},"attestation_state":"computed","paper":{"title":"TransZero: Parallel Tree Expansion in MuZero using Transformer Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Emil Malmsten, Wendelin B\\\"ohmer","submitted_at":"2025-09-14T12:20:38Z","abstract_excerpt":"We present TransZero, a model-based reinforcement learning algorithm that removes the sequential bottleneck in Monte Carlo Tree Search (MCTS). Unlike MuZero, which constructs its search tree step by step using a recurrent dynamics model, TransZero employs a transformer-based network to generate multiple latent future states simultaneously. Combined with the Mean-Variance Constrained (MVC) evaluator that eliminates dependence on inherently sequential visitation counts, our approach enables the parallel expansion of entire subtrees during planning. Experiments in MiniGrid and LunarLander show th"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2509.11233","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2025-09-14T12:20:38Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"0fa3eb6afe3f9d1f7f4c7b37d7859b96f515e584758020b4039b5be10f9227c1","abstract_canon_sha256":"4c9a3be1a6d75ae4049ada4694ec78f958a8e6718f1f26fa81aa8555de472d5a"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T12:11:35.772589Z","signature_b64":"Kp8Ec0yRCtAPUp61wIFAW3d5P+DmmcnMnHFVfkli53cVUPvOTQnA0/CyIiAHt+4T8B2y33/fbzk7tNZQ5ClgDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"553832eccda7f25b144cd5cd52f6f10ac8bcb91531144a925a75faea6fe46a94","last_reissued_at":"2026-07-05T12:11:35.771996Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T12:11:35.771996Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"TransZero: Parallel Tree Expansion in MuZero using Transformer Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Emil Malmsten, Wendelin B\\\"ohmer","submitted_at":"2025-09-14T12:20:38Z","abstract_excerpt":"We present TransZero, a model-based reinforcement learning algorithm that removes the sequential bottleneck in Monte Carlo Tree Search (MCTS). Unlike MuZero, which constructs its search tree step by step using a recurrent dynamics model, TransZero employs a transformer-based network to generate multiple latent future states simultaneously. Combined with the Mean-Variance Constrained (MVC) evaluator that eliminates dependence on inherently sequential visitation counts, our approach enables the parallel expansion of entire subtrees during planning. Experiments in MiniGrid and LunarLander show th"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2509.11233","kind":"arxiv","version":1},"verdict":{"id":null,"model_set":{},"created_at":null,"strongest_claim":"","one_line_summary":"","pipeline_version":null,"weakest_assumption":"","pith_extraction_headline":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2509.11233/integrity.json","findings":[],"available":true,"detectors_run":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938"},"references":{"count":0,"sample":[],"resolved_work":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","internal_anchors":0},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"},"aliases":[{"alias_kind":"arxiv","alias_value":"2509.11233","created_at":"2026-07-05T12:11:35.772056+00:00"},{"alias_kind":"arxiv_version","alias_value":"2509.11233v1","created_at":"2026-07-05T12:11:35.772056+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2509.11233","created_at":"2026-07-05T12:11:35.772056+00:00"},{"alias_kind":"pith_short_12","alias_value":"KU4DF3GNU7ZF","created_at":"2026-07-05T12:11:35.772056+00:00"},{"alias_kind":"pith_short_16","alias_value":"KU4DF3GNU7ZFWFCM","created_at":"2026-07-05T12:11:35.772056+00:00"},{"alias_kind":"pith_short_8","alias_value":"KU4DF3GN","created_at":"2026-07-05T12:11:35.772056+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.08982","citing_title":"PMCTS: Particle Monte Carlo Tree Search for Principled Parallelized Inference Time Scaling","ref_index":61,"is_internal_anchor":false},{"citing_arxiv_id":"2605.08982","citing_title":"PMCTS: Particle Monte Carlo Tree Search for Principled Parallelized Inference Time Scaling","ref_index":62,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/KU4DF3GNU7ZFWFCM2XGVF5XRBL","json":"https://pith.science/pith/KU4DF3GNU7ZFWFCM2XGVF5XRBL.json","graph_json":"https://pith.science/api/pith-number/KU4DF3GNU7ZFWFCM2XGVF5XRBL/graph.json","events_json":"https://pith.science/api/pith-number/KU4DF3GNU7ZFWFCM2XGVF5XRBL/events.json","paper":"https://pith.science/paper/KU4DF3GN"},"agent_actions":{"view_html":"https://pith.science/pith/KU4DF3GNU7ZFWFCM2XGVF5XRBL","download_json":"https://pith.science/pith/KU4DF3GNU7ZFWFCM2XGVF5XRBL.json","view_paper":"https://pith.science/paper/KU4DF3GN","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2509.11233&json=true","fetch_graph":"https://pith.science/api/pith-number/KU4DF3GNU7ZFWFCM2XGVF5XRBL/graph.json","fetch_events":"https://pith.science/api/pith-number/KU4DF3GNU7ZFWFCM2XGVF5XRBL/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/KU4DF3GNU7ZFWFCM2XGVF5XRBL/action/timestamp_anchor","attest_storage":"https://pith.science/pith/KU4DF3GNU7ZFWFCM2XGVF5XRBL/action/storage_attestation","attest_author":"https://pith.science/pith/KU4DF3GNU7ZFWFCM2XGVF5XRBL/action/author_attestation","sign_citation":"https://pith.science/pith/KU4DF3GNU7ZFWFCM2XGVF5XRBL/action/citation_signature","submit_replication":"https://pith.science/pith/KU4DF3GNU7ZFWFCM2XGVF5XRBL/action/replication_record"}},"created_at":"2026-07-05T12:11:35.772056+00:00","updated_at":"2026-07-05T12:11:35.772056+00:00"}