{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2023:HU6KM5L4IMYLXZBV4PQO56QO3C","short_pith_number":"pith:HU6KM5L4","canonical_record":{"source":{"id":"2312.05044","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-08T13:52:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a11f870a3bf6d06b54babb6536783664dc91336ec66dca6e4f83e19252f3e03a","abstract_canon_sha256":"e205fcfc0a3b1508235d76de8bd854e1ca49ed5d40886126cc7c774ce7a46028"},"schema_version":"1.0"},"canonical_sha256":"3d3ca6757c4330bbe435e3e0eefa0ed8aa5b86281a9610fcfcfc419b4b476fb6","source":{"kind":"arxiv","id":"2312.05044","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.05044","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"arxiv_version","alias_value":"2312.05044v2","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.05044","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"pith_short_12","alias_value":"HU6KM5L4IMYL","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"pith_short_16","alias_value":"HU6KM5L4IMYLXZBV","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"pith_short_8","alias_value":"HU6KM5L4","created_at":"2026-07-05T08:08:00Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2023:HU6KM5L4IMYLXZBV4PQO56QO3C","target":"record","payload":{"canonical_record":{"source":{"id":"2312.05044","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-08T13:52:16Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"a11f870a3bf6d06b54babb6536783664dc91336ec66dca6e4f83e19252f3e03a","abstract_canon_sha256":"e205fcfc0a3b1508235d76de8bd854e1ca49ed5d40886126cc7c774ce7a46028"},"schema_version":"1.0"},"canonical_sha256":"3d3ca6757c4330bbe435e3e0eefa0ed8aa5b86281a9610fcfcfc419b4b476fb6","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T08:08:00.377818Z","signature_b64":"9kjh6enCTs/+F0wfSJCei8uL1/ek/54eMMiHbB3YHd/3zsyMP1239LbT/KhO6EbVOqHzPdnNFp1TFFG4jO7dCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3d3ca6757c4330bbe435e3e0eefa0ed8aa5b86281a9610fcfcfc419b4b476fb6","last_reissued_at":"2026-07-05T08:08:00.377395Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T08:08:00.377395Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2312.05044","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:08:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"dio/Vl3knKYLpoAgDYVrZSWV1H8xZjYe/3oeESTwhxd95sc0bCARC1zbG/5JnKnirsYEncl/Hft9iA9MD/koBw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T08:32:13.474102Z"},"content_sha256":"fa8d7c1b0a12b12959dceb386504b1c2506f50fc54ffdc2995307a2e8b347680","schema_version":"1.0","event_id":"sha256:fa8d7c1b0a12b12959dceb386504b1c2506f50fc54ffdc2995307a2e8b347680"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2023:HU6KM5L4IMYLXZBV4PQO56QO3C","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Backward Learning for Goal-Conditioned Policies","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Jan Robine, Marc H\\\"oftmann, Stefan Harmeling","submitted_at":"2023-12-08T13:52:16Z","abstract_excerpt":"Can we learn policies in reinforcement learning without rewards? Can we learn a policy just by trying to reach a goal state? We answer these questions positively by proposing a multi-step procedure that first learns a world model that goes backward in time, secondly generates goal-reaching backward trajectories, thirdly improves those sequences using shortest path finding algorithms, and finally trains a neural network policy by imitation learning. We evaluate our method on a deterministic maze environment where the observations are $64\\times 64$ pixel bird's eye images and can show that it co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.05044","kind":"arxiv","version":2},"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/2312.05044/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T08:08:00Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"sodL41jELe1I8Wcvnb5hyjDhTanK/lFDYFNWO1BwZm/zwxq4FIPjb4t8fjV8BIRweXaKVAdjVsp63wv2Umw4Bg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-01T08:32:13.474459Z"},"content_sha256":"fbe4ba31f680d7d9fcfbcb52b3fe773b94f865281ac994f159eb07e9e6554200","schema_version":"1.0","event_id":"sha256:fbe4ba31f680d7d9fcfbcb52b3fe773b94f865281ac994f159eb07e9e6554200"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/HU6KM5L4IMYLXZBV4PQO56QO3C/bundle.json","state_url":"https://pith.science/pith/HU6KM5L4IMYLXZBV4PQO56QO3C/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/HU6KM5L4IMYLXZBV4PQO56QO3C/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-01T08:32:13Z","links":{"resolver":"https://pith.science/pith/HU6KM5L4IMYLXZBV4PQO56QO3C","bundle":"https://pith.science/pith/HU6KM5L4IMYLXZBV4PQO56QO3C/bundle.json","state":"https://pith.science/pith/HU6KM5L4IMYLXZBV4PQO56QO3C/state.json","well_known_bundle":"https://pith.science/.well-known/pith/HU6KM5L4IMYLXZBV4PQO56QO3C/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2023:HU6KM5L4IMYLXZBV4PQO56QO3C","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"e205fcfc0a3b1508235d76de8bd854e1ca49ed5d40886126cc7c774ce7a46028","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-08T13:52:16Z","title_canon_sha256":"a11f870a3bf6d06b54babb6536783664dc91336ec66dca6e4f83e19252f3e03a"},"schema_version":"1.0","source":{"id":"2312.05044","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2312.05044","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"arxiv_version","alias_value":"2312.05044v2","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2312.05044","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"pith_short_12","alias_value":"HU6KM5L4IMYL","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"pith_short_16","alias_value":"HU6KM5L4IMYLXZBV","created_at":"2026-07-05T08:08:00Z"},{"alias_kind":"pith_short_8","alias_value":"HU6KM5L4","created_at":"2026-07-05T08:08:00Z"}],"graph_snapshots":[{"event_id":"sha256:fbe4ba31f680d7d9fcfbcb52b3fe773b94f865281ac994f159eb07e9e6554200","target":"graph","created_at":"2026-07-05T08:08:00Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2312.05044/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Can we learn policies in reinforcement learning without rewards? Can we learn a policy just by trying to reach a goal state? We answer these questions positively by proposing a multi-step procedure that first learns a world model that goes backward in time, secondly generates goal-reaching backward trajectories, thirdly improves those sequences using shortest path finding algorithms, and finally trains a neural network policy by imitation learning. We evaluate our method on a deterministic maze environment where the observations are $64\\times 64$ pixel bird's eye images and can show that it co","authors_text":"Jan Robine, Marc H\\\"oftmann, Stefan Harmeling","cross_cats":["cs.AI"],"headline":"","license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-08T13:52:16Z","title":"Backward Learning for Goal-Conditioned Policies"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2312.05044","kind":"arxiv","version":2},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:fa8d7c1b0a12b12959dceb386504b1c2506f50fc54ffdc2995307a2e8b347680","target":"record","created_at":"2026-07-05T08:08:00Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"e205fcfc0a3b1508235d76de8bd854e1ca49ed5d40886126cc7c774ce7a46028","cross_cats_sorted":["cs.AI"],"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-12-08T13:52:16Z","title_canon_sha256":"a11f870a3bf6d06b54babb6536783664dc91336ec66dca6e4f83e19252f3e03a"},"schema_version":"1.0","source":{"id":"2312.05044","kind":"arxiv","version":2}},"canonical_sha256":"3d3ca6757c4330bbe435e3e0eefa0ed8aa5b86281a9610fcfcfc419b4b476fb6","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"3d3ca6757c4330bbe435e3e0eefa0ed8aa5b86281a9610fcfcfc419b4b476fb6","first_computed_at":"2026-07-05T08:08:00.377395Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T08:08:00.377395Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9kjh6enCTs/+F0wfSJCei8uL1/ek/54eMMiHbB3YHd/3zsyMP1239LbT/KhO6EbVOqHzPdnNFp1TFFG4jO7dCw==","signature_status":"signed_v1","signed_at":"2026-07-05T08:08:00.377818Z","signed_message":"canonical_sha256_bytes"},"source_id":"2312.05044","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:fa8d7c1b0a12b12959dceb386504b1c2506f50fc54ffdc2995307a2e8b347680","sha256:fbe4ba31f680d7d9fcfbcb52b3fe773b94f865281ac994f159eb07e9e6554200"],"state_sha256":"f6c036fa53739a40ec6ed346f27d06d1de570c1eafaed6ce64ae44b1ec9b6e58"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"H34/wJjSTOamIlcoMKAcl2DG+TPzOdihDQHAMbyL5MAu0vxKkwc1uFUd7xFLEW6cWbiIhGoVlLcquOWtUhDHCg==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-01T08:32:13.478619Z","bundle_sha256":"1e0d334a8f1d71c2f635ed70d6f5ef62dde5f591314108ba6c0b2c17e1b2d412"}}