{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:BRDCFOTIKDOLNOBL7RZQJMHOIK","short_pith_number":"pith:BRDCFOTI","schema_version":"1.0","canonical_sha256":"0c4622ba6850dcb6b82bfc7304b0ee428e70151fa032b70a494bac552821abcd","source":{"kind":"arxiv","id":"1906.02736","version":1},"attestation_state":"computed","paper":{"title":"DeepMDP: Learning Continuous Latent Space Models for Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Carles Gelada, Jacob Buckman, Marc G. Bellemare, Ofir Nachum, Saurabh Kumar","submitted_at":"2019-06-06T17:55:17Z","abstract_excerpt":"Many reinforcement learning (RL) tasks provide the agent with high-dimensional observations that can be simplified into low-dimensional continuous states. To formalize this process, we introduce the concept of a DeepMDP, a parameterized latent space model that is trained via the minimization of two tractable losses: prediction of rewards and prediction of the distribution over next latent states. We show that the optimization of these objectives guarantees (1) the quality of the latent space as a representation of the state space and (2) the quality of the DeepMDP as a model of the environment"},"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":"1906.02736","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-06-06T17:55:17Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"1ade1b596c52b9536d90b041353ccb6a28f8a1e8e3ba4a21c21e82c325d18792","abstract_canon_sha256":"ca6381aa95f8df0ab8c6c9d758f23cf4a93cacd51478cc583daa9a707b9524b2"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:44:00.364266Z","signature_b64":"0/epbeVNbm5PZbAEQ/5Keoovm8nOc86juTZmqHX7zydqJJr5a6Smw9ZI1YvBCxZY1VHaqmBq0WDKSc7zrF32Cw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"0c4622ba6850dcb6b82bfc7304b0ee428e70151fa032b70a494bac552821abcd","last_reissued_at":"2026-05-17T23:44:00.363750Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:44:00.363750Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"DeepMDP: Learning Continuous Latent Space Models for Representation Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Carles Gelada, Jacob Buckman, Marc G. Bellemare, Ofir Nachum, Saurabh Kumar","submitted_at":"2019-06-06T17:55:17Z","abstract_excerpt":"Many reinforcement learning (RL) tasks provide the agent with high-dimensional observations that can be simplified into low-dimensional continuous states. To formalize this process, we introduce the concept of a DeepMDP, a parameterized latent space model that is trained via the minimization of two tractable losses: prediction of rewards and prediction of the distribution over next latent states. We show that the optimization of these objectives guarantees (1) the quality of the latent space as a representation of the state space and (2) the quality of the DeepMDP as a model of the environment"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1906.02736","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":""},"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":"1906.02736","created_at":"2026-05-17T23:44:00.363844+00:00"},{"alias_kind":"arxiv_version","alias_value":"1906.02736v1","created_at":"2026-05-17T23:44:00.363844+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1906.02736","created_at":"2026-05-17T23:44:00.363844+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2203.12601","citing_title":"R3M: A Universal Visual Representation for Robot Manipulation","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2605.09364","citing_title":"Multi-scale Predictive Representations for Goal-conditioned Reinforcement Learning","ref_index":12,"is_internal_anchor":false},{"citing_arxiv_id":"1912.01603","citing_title":"Dream to Control: Learning Behaviors by Latent Imagination","ref_index":15,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/BRDCFOTIKDOLNOBL7RZQJMHOIK","json":"https://pith.science/pith/BRDCFOTIKDOLNOBL7RZQJMHOIK.json","graph_json":"https://pith.science/api/pith-number/BRDCFOTIKDOLNOBL7RZQJMHOIK/graph.json","events_json":"https://pith.science/api/pith-number/BRDCFOTIKDOLNOBL7RZQJMHOIK/events.json","paper":"https://pith.science/paper/BRDCFOTI"},"agent_actions":{"view_html":"https://pith.science/pith/BRDCFOTIKDOLNOBL7RZQJMHOIK","download_json":"https://pith.science/pith/BRDCFOTIKDOLNOBL7RZQJMHOIK.json","view_paper":"https://pith.science/paper/BRDCFOTI","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1906.02736&json=true","fetch_graph":"https://pith.science/api/pith-number/BRDCFOTIKDOLNOBL7RZQJMHOIK/graph.json","fetch_events":"https://pith.science/api/pith-number/BRDCFOTIKDOLNOBL7RZQJMHOIK/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/BRDCFOTIKDOLNOBL7RZQJMHOIK/action/timestamp_anchor","attest_storage":"https://pith.science/pith/BRDCFOTIKDOLNOBL7RZQJMHOIK/action/storage_attestation","attest_author":"https://pith.science/pith/BRDCFOTIKDOLNOBL7RZQJMHOIK/action/author_attestation","sign_citation":"https://pith.science/pith/BRDCFOTIKDOLNOBL7RZQJMHOIK/action/citation_signature","submit_replication":"https://pith.science/pith/BRDCFOTIKDOLNOBL7RZQJMHOIK/action/replication_record"}},"created_at":"2026-05-17T23:44:00.363844+00:00","updated_at":"2026-05-17T23:44:00.363844+00:00"}