{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:4H5XE4S3KRTALLQUIGAELWXYRV","short_pith_number":"pith:4H5XE4S3","schema_version":"1.0","canonical_sha256":"e1fb72725b546605ae14418045daf88d476a2a393e9bb770866faf404ef7799b","source":{"kind":"arxiv","id":"1702.08360","version":1},"attestation_state":"computed","paper":{"title":"Neural Map: Structured Memory for Deep Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Emilio Parisotto, Ruslan Salakhutdinov","submitted_at":"2017-02-27T16:32:27Z","abstract_excerpt":"A critical component to enabling intelligent reasoning in partially observable environments is memory. Despite this importance, Deep Reinforcement Learning (DRL) agents have so far used relatively simple memory architectures, with the main methods to overcome partial observability being either a temporal convolution over the past k frames or an LSTM layer. More recent work (Oh et al., 2016) has went beyond these architectures by using memory networks which can allow more sophisticated addressing schemes over the past k frames. But even these architectures are unsatisfactory due to the reason t"},"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":"1702.08360","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-02-27T16:32:27Z","cross_cats_sorted":[],"title_canon_sha256":"9c577101d480e721b2a6eacda0e5363ede3bae42eef1ce22b128cf570ba45336","abstract_canon_sha256":"14ba9381ba83295bf71543b89014bc2f10aa1e2f845da7a28712d55bb5ea1150"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:49:55.122585Z","signature_b64":"h5LmRQaHdqIV+6ktCE5MY6dDJZ3xL/W+wC4Z0GJC0hY+ZRALqy2s6bWhAh2/tKEnesds9jjcKdGx36t6XEbkAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"e1fb72725b546605ae14418045daf88d476a2a393e9bb770866faf404ef7799b","last_reissued_at":"2026-05-18T00:49:55.121853Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:49:55.121853Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Neural Map: Structured Memory for Deep Reinforcement Learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Emilio Parisotto, Ruslan Salakhutdinov","submitted_at":"2017-02-27T16:32:27Z","abstract_excerpt":"A critical component to enabling intelligent reasoning in partially observable environments is memory. Despite this importance, Deep Reinforcement Learning (DRL) agents have so far used relatively simple memory architectures, with the main methods to overcome partial observability being either a temporal convolution over the past k frames or an LSTM layer. More recent work (Oh et al., 2016) has went beyond these architectures by using memory networks which can allow more sophisticated addressing schemes over the past k frames. But even these architectures are unsatisfactory due to the reason t"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1702.08360","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":"1702.08360","created_at":"2026-05-18T00:49:55.121982+00:00"},{"alias_kind":"arxiv_version","alias_value":"1702.08360v1","created_at":"2026-05-18T00:49:55.121982+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1702.08360","created_at":"2026-05-18T00:49:55.121982+00:00"},{"alias_kind":"pith_short_12","alias_value":"4H5XE4S3KRTA","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_16","alias_value":"4H5XE4S3KRTALLQU","created_at":"2026-05-18T12:30:58.224056+00:00"},{"alias_kind":"pith_short_8","alias_value":"4H5XE4S3","created_at":"2026-05-18T12:30:58.224056+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":5,"internal_anchor_count":5,"sample":[{"citing_arxiv_id":"2606.10401","citing_title":"CoCoSI: Collaborative Cognitive Map Construction for Spatial Intelligence","ref_index":15,"is_internal_anchor":true},{"citing_arxiv_id":"2605.27686","citing_title":"Tensor Memory: Fixed-Size Recurrent State for Long-Horizon Transformers","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2605.31404","citing_title":"The Sword, Shield, and Achilles' Heel: Characterizing the Linguistic Inductive Bias of Large Language Models for Spatial Reasoning in Navigation Planning","ref_index":22,"is_internal_anchor":true},{"citing_arxiv_id":"1907.11770","citing_title":"To Learn or Not to Learn: Analyzing the Role of Learning for Navigation in Virtual Environments","ref_index":26,"is_internal_anchor":true},{"citing_arxiv_id":"2604.16331","citing_title":"BrainMem: Brain-Inspired Evolving Memory for Embodied Agent Task Planning","ref_index":30,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/4H5XE4S3KRTALLQUIGAELWXYRV","json":"https://pith.science/pith/4H5XE4S3KRTALLQUIGAELWXYRV.json","graph_json":"https://pith.science/api/pith-number/4H5XE4S3KRTALLQUIGAELWXYRV/graph.json","events_json":"https://pith.science/api/pith-number/4H5XE4S3KRTALLQUIGAELWXYRV/events.json","paper":"https://pith.science/paper/4H5XE4S3"},"agent_actions":{"view_html":"https://pith.science/pith/4H5XE4S3KRTALLQUIGAELWXYRV","download_json":"https://pith.science/pith/4H5XE4S3KRTALLQUIGAELWXYRV.json","view_paper":"https://pith.science/paper/4H5XE4S3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1702.08360&json=true","fetch_graph":"https://pith.science/api/pith-number/4H5XE4S3KRTALLQUIGAELWXYRV/graph.json","fetch_events":"https://pith.science/api/pith-number/4H5XE4S3KRTALLQUIGAELWXYRV/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/4H5XE4S3KRTALLQUIGAELWXYRV/action/timestamp_anchor","attest_storage":"https://pith.science/pith/4H5XE4S3KRTALLQUIGAELWXYRV/action/storage_attestation","attest_author":"https://pith.science/pith/4H5XE4S3KRTALLQUIGAELWXYRV/action/author_attestation","sign_citation":"https://pith.science/pith/4H5XE4S3KRTALLQUIGAELWXYRV/action/citation_signature","submit_replication":"https://pith.science/pith/4H5XE4S3KRTALLQUIGAELWXYRV/action/replication_record"}},"created_at":"2026-05-18T00:49:55.121982+00:00","updated_at":"2026-05-18T00:49:55.121982+00:00"}