{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:NPXYFX5JCHROY56G6YAUMNLXOK","short_pith_number":"pith:NPXYFX5J","canonical_record":{"source":{"id":"1902.04706","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-13T02:14:13Z","cross_cats_sorted":["cs.RO","stat.ML"],"title_canon_sha256":"a848395bddfc6dbba8161b9affc932bd37e5938b45e893f1ccd2b4ef81daf7e4","abstract_canon_sha256":"79a087ab438d3f174b09800622a7a5871032401e965f6678cb80db28ecb83ace"},"schema_version":"1.0"},"canonical_sha256":"6bef82dfa911e2ec77c6f60146357772b3b45e3c03e271538a6a2afd0b8e6a90","source":{"kind":"arxiv","id":"1902.04706","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1902.04706","created_at":"2026-05-17T23:53:40Z"},{"alias_kind":"arxiv_version","alias_value":"1902.04706v2","created_at":"2026-05-17T23:53:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.04706","created_at":"2026-05-17T23:53:40Z"},{"alias_kind":"pith_short_12","alias_value":"NPXYFX5JCHRO","created_at":"2026-05-18T12:33:24Z"},{"alias_kind":"pith_short_16","alias_value":"NPXYFX5JCHROY56G","created_at":"2026-05-18T12:33:24Z"},{"alias_kind":"pith_short_8","alias_value":"NPXYFX5J","created_at":"2026-05-18T12:33:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:NPXYFX5JCHROY56G6YAUMNLXOK","target":"record","payload":{"canonical_record":{"source":{"id":"1902.04706","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-13T02:14:13Z","cross_cats_sorted":["cs.RO","stat.ML"],"title_canon_sha256":"a848395bddfc6dbba8161b9affc932bd37e5938b45e893f1ccd2b4ef81daf7e4","abstract_canon_sha256":"79a087ab438d3f174b09800622a7a5871032401e965f6678cb80db28ecb83ace"},"schema_version":"1.0"},"canonical_sha256":"6bef82dfa911e2ec77c6f60146357772b3b45e3c03e271538a6a2afd0b8e6a90","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:53:40.973693Z","signature_b64":"EzHHOjZuUD4wXDx5e/xKBi96sai1IHb0irxuyd/4MtiFcy96uHKEMxHmKe6Yby82AwE+wH8R+R2d2jKVL/GBDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"6bef82dfa911e2ec77c6f60146357772b3b45e3c03e271538a6a2afd0b8e6a90","last_reissued_at":"2026-05-17T23:53:40.973201Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:53:40.973201Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1902.04706","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":"2026-05-17T23:53:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"acQGNFFscEAxwc7kICfb3cZII+awwO/MosZR6McHhSC5KfiVYmk/WFIW972OT8wf7764jJRRAEwJ6nYfHarlCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T17:55:15.321714Z"},"content_sha256":"dfb4f1114f36cc555dc516a002e007aacf0bc5203c79681b16318a83513397fd","schema_version":"1.0","event_id":"sha256:dfb4f1114f36cc555dc516a002e007aacf0bc5203c79681b16318a83513397fd"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:NPXYFX5JCHROY56G6YAUMNLXOK","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Simultaneously Learning Vision and Feature-based Control Policies for Real-world Ball-in-a-Cup","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.RO","stat.ML"],"primary_cat":"cs.LG","authors_text":"Abbas Abdolmaleki, Devin Schwab, Francesco Nori, Martin Riedmiller, Michael Neunert, Murilo F. Martins, Roland Hafner, Thomas Lampe, Tim Hertweck, Tobias Springenberg","submitted_at":"2019-02-13T02:14:13Z","abstract_excerpt":"We present a method for fast training of vision based control policies on real robots. The key idea behind our method is to perform multi-task Reinforcement Learning with auxiliary tasks that differ not only in the reward to be optimized but also in the state-space in which they operate. In particular, we allow auxiliary task policies to utilize task features that are available only at training-time. This allows for fast learning of auxiliary policies, which subsequently generate good data for training the main, vision-based control policies. This method can be seen as an extension of the Sche"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.04706","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":""},"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":"2026-05-17T23:53:40Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"j8sLSUBNfVK+ZKLxTzCm2szeWNHlI2TI7C1Afc9650Qu5FlUr12kuVd2f5PXpaNwtvB+EmeqABkf6IM1n2opDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-26T17:55:15.322354Z"},"content_sha256":"6a9ac8863bf63406f0c3deb78a62bf2d28278e91442431fc6e35d521ae6259ee","schema_version":"1.0","event_id":"sha256:6a9ac8863bf63406f0c3deb78a62bf2d28278e91442431fc6e35d521ae6259ee"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/NPXYFX5JCHROY56G6YAUMNLXOK/bundle.json","state_url":"https://pith.science/pith/NPXYFX5JCHROY56G6YAUMNLXOK/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/NPXYFX5JCHROY56G6YAUMNLXOK/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":"2026-05-26T17:55:15Z","links":{"resolver":"https://pith.science/pith/NPXYFX5JCHROY56G6YAUMNLXOK","bundle":"https://pith.science/pith/NPXYFX5JCHROY56G6YAUMNLXOK/bundle.json","state":"https://pith.science/pith/NPXYFX5JCHROY56G6YAUMNLXOK/state.json","well_known_bundle":"https://pith.science/.well-known/pith/NPXYFX5JCHROY56G6YAUMNLXOK/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:NPXYFX5JCHROY56G6YAUMNLXOK","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":"79a087ab438d3f174b09800622a7a5871032401e965f6678cb80db28ecb83ace","cross_cats_sorted":["cs.RO","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-13T02:14:13Z","title_canon_sha256":"a848395bddfc6dbba8161b9affc932bd37e5938b45e893f1ccd2b4ef81daf7e4"},"schema_version":"1.0","source":{"id":"1902.04706","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1902.04706","created_at":"2026-05-17T23:53:40Z"},{"alias_kind":"arxiv_version","alias_value":"1902.04706v2","created_at":"2026-05-17T23:53:40Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.04706","created_at":"2026-05-17T23:53:40Z"},{"alias_kind":"pith_short_12","alias_value":"NPXYFX5JCHRO","created_at":"2026-05-18T12:33:24Z"},{"alias_kind":"pith_short_16","alias_value":"NPXYFX5JCHROY56G","created_at":"2026-05-18T12:33:24Z"},{"alias_kind":"pith_short_8","alias_value":"NPXYFX5J","created_at":"2026-05-18T12:33:24Z"}],"graph_snapshots":[{"event_id":"sha256:6a9ac8863bf63406f0c3deb78a62bf2d28278e91442431fc6e35d521ae6259ee","target":"graph","created_at":"2026-05-17T23:53:40Z","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"},"paper":{"abstract_excerpt":"We present a method for fast training of vision based control policies on real robots. The key idea behind our method is to perform multi-task Reinforcement Learning with auxiliary tasks that differ not only in the reward to be optimized but also in the state-space in which they operate. In particular, we allow auxiliary task policies to utilize task features that are available only at training-time. This allows for fast learning of auxiliary policies, which subsequently generate good data for training the main, vision-based control policies. This method can be seen as an extension of the Sche","authors_text":"Abbas Abdolmaleki, Devin Schwab, Francesco Nori, Martin Riedmiller, Michael Neunert, Murilo F. Martins, Roland Hafner, Thomas Lampe, Tim Hertweck, Tobias Springenberg","cross_cats":["cs.RO","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-13T02:14:13Z","title":"Simultaneously Learning Vision and Feature-based Control Policies for Real-world Ball-in-a-Cup"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.04706","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:dfb4f1114f36cc555dc516a002e007aacf0bc5203c79681b16318a83513397fd","target":"record","created_at":"2026-05-17T23:53:40Z","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":"79a087ab438d3f174b09800622a7a5871032401e965f6678cb80db28ecb83ace","cross_cats_sorted":["cs.RO","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-13T02:14:13Z","title_canon_sha256":"a848395bddfc6dbba8161b9affc932bd37e5938b45e893f1ccd2b4ef81daf7e4"},"schema_version":"1.0","source":{"id":"1902.04706","kind":"arxiv","version":2}},"canonical_sha256":"6bef82dfa911e2ec77c6f60146357772b3b45e3c03e271538a6a2afd0b8e6a90","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"6bef82dfa911e2ec77c6f60146357772b3b45e3c03e271538a6a2afd0b8e6a90","first_computed_at":"2026-05-17T23:53:40.973201Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:53:40.973201Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"EzHHOjZuUD4wXDx5e/xKBi96sai1IHb0irxuyd/4MtiFcy96uHKEMxHmKe6Yby82AwE+wH8R+R2d2jKVL/GBDQ==","signature_status":"signed_v1","signed_at":"2026-05-17T23:53:40.973693Z","signed_message":"canonical_sha256_bytes"},"source_id":"1902.04706","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:dfb4f1114f36cc555dc516a002e007aacf0bc5203c79681b16318a83513397fd","sha256:6a9ac8863bf63406f0c3deb78a62bf2d28278e91442431fc6e35d521ae6259ee"],"state_sha256":"610ea21fb0cf689cc74c3ce7b9289e112e4085ce2b98e02592d0ade9a797148b"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"/Q2QnY9AduZF7Txf7QX2lKBIl4hNdSvREApN5EDwzxp9vRTGPypjYdM9wTC+1PIVxbDkikhJ5hU+B3Ala7/cDQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-26T17:55:15.325762Z","bundle_sha256":"0a48361c0c7bbf029598cf64321bae266bfea2cdd60d3133418b233648ad2bec"}}