{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:3ZWCR3BZI2ULGKUNTB2WXUYWI6","short_pith_number":"pith:3ZWCR3BZ","schema_version":"1.0","canonical_sha256":"de6c28ec3946a8b32a8d98756bd31647bb2fa25186e459ef220b25b8c5f462ef","source":{"kind":"arxiv","id":"1906.08226","version":6},"attestation_state":"computed","paper":{"title":"Unsupervised State Representation Learning in Atari","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ankesh Anand, Evan Racah, Marc-Alexandre C\\^ot\\'e, R Devon Hjelm, Sherjil Ozair, Yoshua Bengio","submitted_at":"2019-06-19T17:16:46Z","abstract_excerpt":"State representation learning, or the ability to capture latent generative factors of an environment, is crucial for building intelligent agents that can perform a wide variety of tasks. Learning such representations without supervision from rewards is a challenging open problem. We introduce a method that learns state representations by maximizing mutual information across spatially and temporally distinct features of a neural encoder of the observations. We also introduce a new benchmark based on Atari 2600 games where we evaluate representations based on how well they capture the ground tru"},"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":"1906.08226","kind":"arxiv","version":6},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-19T17:16:46Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"c8b701c496f09490722a2db32a0d62181cb191d14fab05aae592cec06876e452","abstract_canon_sha256":"47b4bd7bf1509f75bee06e0fe1e857cd6cd16a5d0b21e08e2b931fd01ceacdc0"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:49:34.888098Z","signature_b64":"4ZIWgKd+7n4VJRpJyIsqtsFgPzhjkswmmNUgRBdhZelW0p007ck7hQNWhO0jEMi83gERk+kY0HPKLahrF4IqBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"de6c28ec3946a8b32a8d98756bd31647bb2fa25186e459ef220b25b8c5f462ef","last_reissued_at":"2026-07-05T01:49:34.887559Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:49:34.887559Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Unsupervised State Representation Learning in Atari","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Ankesh Anand, Evan Racah, Marc-Alexandre C\\^ot\\'e, R Devon Hjelm, Sherjil Ozair, Yoshua Bengio","submitted_at":"2019-06-19T17:16:46Z","abstract_excerpt":"State representation learning, or the ability to capture latent generative factors of an environment, is crucial for building intelligent agents that can perform a wide variety of tasks. Learning such representations without supervision from rewards is a challenging open problem. We introduce a method that learns state representations by maximizing mutual information across spatially and temporally distinct features of a neural encoder of the observations. We also introduce a new benchmark based on Atari 2600 games where we evaluate representations based on how well they capture the ground tru"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.08226","kind":"arxiv","version":6},"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/1906.08226/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":"1906.08226","created_at":"2026-07-05T01:49:34.887617+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.08226v6","created_at":"2026-07-05T01:49:34.887617+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.08226","created_at":"2026-07-05T01:49:34.887617+00:00"},{"alias_kind":"pith_short_12","alias_value":"3ZWCR3BZI2UL","created_at":"2026-07-05T01:49:34.887617+00:00"},{"alias_kind":"pith_short_16","alias_value":"3ZWCR3BZI2ULGKUN","created_at":"2026-07-05T01:49:34.887617+00:00"},{"alias_kind":"pith_short_8","alias_value":"3ZWCR3BZ","created_at":"2026-07-05T01:49:34.887617+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.17518","citing_title":"A Survey of State Representation Learning for Deep Reinforcement Learning","ref_index":8,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/3ZWCR3BZI2ULGKUNTB2WXUYWI6","json":"https://pith.science/pith/3ZWCR3BZI2ULGKUNTB2WXUYWI6.json","graph_json":"https://pith.science/api/pith-number/3ZWCR3BZI2ULGKUNTB2WXUYWI6/graph.json","events_json":"https://pith.science/api/pith-number/3ZWCR3BZI2ULGKUNTB2WXUYWI6/events.json","paper":"https://pith.science/paper/3ZWCR3BZ"},"agent_actions":{"view_html":"https://pith.science/pith/3ZWCR3BZI2ULGKUNTB2WXUYWI6","download_json":"https://pith.science/pith/3ZWCR3BZI2ULGKUNTB2WXUYWI6.json","view_paper":"https://pith.science/paper/3ZWCR3BZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.08226&json=true","fetch_graph":"https://pith.science/api/pith-number/3ZWCR3BZI2ULGKUNTB2WXUYWI6/graph.json","fetch_events":"https://pith.science/api/pith-number/3ZWCR3BZI2ULGKUNTB2WXUYWI6/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/3ZWCR3BZI2ULGKUNTB2WXUYWI6/action/timestamp_anchor","attest_storage":"https://pith.science/pith/3ZWCR3BZI2ULGKUNTB2WXUYWI6/action/storage_attestation","attest_author":"https://pith.science/pith/3ZWCR3BZI2ULGKUNTB2WXUYWI6/action/author_attestation","sign_citation":"https://pith.science/pith/3ZWCR3BZI2ULGKUNTB2WXUYWI6/action/citation_signature","submit_replication":"https://pith.science/pith/3ZWCR3BZI2ULGKUNTB2WXUYWI6/action/replication_record"}},"created_at":"2026-07-05T01:49:34.887617+00:00","updated_at":"2026-07-05T01:49:34.887617+00:00"}