{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:K7SMXBUONCYS7EC2SZZ4OOTUNG","short_pith_number":"pith:K7SMXBUO","schema_version":"1.0","canonical_sha256":"57e4cb868e68b12f905a9673c73a7469ad069946d343df96e8112d40230536bd","source":{"kind":"arxiv","id":"2304.14698","version":1},"attestation_state":"computed","paper":{"title":"X-RLflow: Graph Reinforcement Learning for Neural Network Subgraphs Transformation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Eiko Yoneki, Guoliang He, Sean Parker","submitted_at":"2023-04-28T09:06:18Z","abstract_excerpt":"Tensor graph superoptimisation systems perform a sequence of subgraph substitution to neural networks, to find the optimal computation graph structure. Such a graph transformation process naturally falls into the framework of sequential decision-making, and existing systems typically employ a greedy search approach, which cannot explore the whole search space as it cannot tolerate a temporary loss of performance. In this paper, we address the tensor graph superoptimisation problem by exploring an alternative search approach, reinforcement learning (RL). Our proposed approach, X-RLflow, can lea"},"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":"2304.14698","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2023-04-28T09:06:18Z","cross_cats_sorted":["cs.AI"],"title_canon_sha256":"aacdc8aa1d8687b17d022baaf9e7af3ff6acbda1fb4cf6f8d352b783a3d65e54","abstract_canon_sha256":"b2134f2fe3ff2248cba0dfb53e991165e2fcba3ea273566c6a36b126d7845db6"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:05:20.184720Z","signature_b64":"Vw/WwNMoY9GiUcRiWm/Bm7bmZ6DhRP/PgqFxp1m89PCzGwQVdFSZhWNoXDVc9v/I71RcdHkdEQh+IpA54YE7BA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"57e4cb868e68b12f905a9673c73a7469ad069946d343df96e8112d40230536bd","last_reissued_at":"2026-07-05T06:05:20.184299Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:05:20.184299Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"X-RLflow: Graph Reinforcement Learning for Neural Network Subgraphs Transformation","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI"],"primary_cat":"cs.LG","authors_text":"Eiko Yoneki, Guoliang He, Sean Parker","submitted_at":"2023-04-28T09:06:18Z","abstract_excerpt":"Tensor graph superoptimisation systems perform a sequence of subgraph substitution to neural networks, to find the optimal computation graph structure. Such a graph transformation process naturally falls into the framework of sequential decision-making, and existing systems typically employ a greedy search approach, which cannot explore the whole search space as it cannot tolerate a temporary loss of performance. In this paper, we address the tensor graph superoptimisation problem by exploring an alternative search approach, reinforcement learning (RL). Our proposed approach, X-RLflow, can lea"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2304.14698","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2304.14698/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":"2304.14698","created_at":"2026-07-05T06:05:20.184372+00:00"},{"alias_kind":"arxiv_version","alias_value":"2304.14698v1","created_at":"2026-07-05T06:05:20.184372+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2304.14698","created_at":"2026-07-05T06:05:20.184372+00:00"},{"alias_kind":"pith_short_12","alias_value":"K7SMXBUONCYS","created_at":"2026-07-05T06:05:20.184372+00:00"},{"alias_kind":"pith_short_16","alias_value":"K7SMXBUONCYS7EC2","created_at":"2026-07-05T06:05:20.184372+00:00"},{"alias_kind":"pith_short_8","alias_value":"K7SMXBUO","created_at":"2026-07-05T06:05:20.184372+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2506.01880","citing_title":"Pearl: Automatic Code Optimization Using Deep Reinforcement Learning","ref_index":28,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K7SMXBUONCYS7EC2SZZ4OOTUNG","json":"https://pith.science/pith/K7SMXBUONCYS7EC2SZZ4OOTUNG.json","graph_json":"https://pith.science/api/pith-number/K7SMXBUONCYS7EC2SZZ4OOTUNG/graph.json","events_json":"https://pith.science/api/pith-number/K7SMXBUONCYS7EC2SZZ4OOTUNG/events.json","paper":"https://pith.science/paper/K7SMXBUO"},"agent_actions":{"view_html":"https://pith.science/pith/K7SMXBUONCYS7EC2SZZ4OOTUNG","download_json":"https://pith.science/pith/K7SMXBUONCYS7EC2SZZ4OOTUNG.json","view_paper":"https://pith.science/paper/K7SMXBUO","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2304.14698&json=true","fetch_graph":"https://pith.science/api/pith-number/K7SMXBUONCYS7EC2SZZ4OOTUNG/graph.json","fetch_events":"https://pith.science/api/pith-number/K7SMXBUONCYS7EC2SZZ4OOTUNG/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K7SMXBUONCYS7EC2SZZ4OOTUNG/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K7SMXBUONCYS7EC2SZZ4OOTUNG/action/storage_attestation","attest_author":"https://pith.science/pith/K7SMXBUONCYS7EC2SZZ4OOTUNG/action/author_attestation","sign_citation":"https://pith.science/pith/K7SMXBUONCYS7EC2SZZ4OOTUNG/action/citation_signature","submit_replication":"https://pith.science/pith/K7SMXBUONCYS7EC2SZZ4OOTUNG/action/replication_record"}},"created_at":"2026-07-05T06:05:20.184372+00:00","updated_at":"2026-07-05T06:05:20.184372+00:00"}