{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:MIVZPVIPGPVXALAGHPEIQPZXMO","short_pith_number":"pith:MIVZPVIP","schema_version":"1.0","canonical_sha256":"622b97d50f33eb702c063bc8883f3763bdd18efc111ae18c82c99190323d686e","source":{"kind":"arxiv","id":"2501.08455","version":1},"attestation_state":"computed","paper":{"title":"Keras Sig: Efficient Path Signature Computation on GPU in Keras 3","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Hugo Inzirillo, R\\'emi Genet","submitted_at":"2025-01-14T22:00:01Z","abstract_excerpt":"In this paper we introduce Keras Sig a high-performance pythonic library designed to compute path signature for deep learning applications. Entirely built in Keras 3, \\textit{Keras Sig} leverages the seamless integration with the mostly used deep learning backends such as PyTorch, JAX and TensorFlow. Inspired by Kidger and Lyons (2021),we proposed a novel approach reshaping signature calculations to leverage GPU parallelism. This adjustment allows us to reduce the training time by 55\\% and 5 to 10-fold improvements in direct signature computation compared to existing methods, while maintaining"},"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":"2501.08455","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2025-01-14T22:00:01Z","cross_cats_sorted":["cs.DC"],"title_canon_sha256":"1a10331040559837cf7cfba855c9135386bd475d933c2a0bd7ba84dc1a4a5d1a","abstract_canon_sha256":"d5e98a6be73f2261d056646816e916753683124776d209fd6edfeb8dda580937"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T10:01:20.205060Z","signature_b64":"KEFnMaHn3qvdoHw7uWDvr773JbMBB565hd1A19vSjZOcSw9xhVzpHI9jW12WyrdsUCq34t2Y8D0JMyGko1gvCA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"622b97d50f33eb702c063bc8883f3763bdd18efc111ae18c82c99190323d686e","last_reissued_at":"2026-07-05T10:01:20.204658Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T10:01:20.204658Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Keras Sig: Efficient Path Signature Computation on GPU in Keras 3","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.DC"],"primary_cat":"cs.LG","authors_text":"Hugo Inzirillo, R\\'emi Genet","submitted_at":"2025-01-14T22:00:01Z","abstract_excerpt":"In this paper we introduce Keras Sig a high-performance pythonic library designed to compute path signature for deep learning applications. Entirely built in Keras 3, \\textit{Keras Sig} leverages the seamless integration with the mostly used deep learning backends such as PyTorch, JAX and TensorFlow. Inspired by Kidger and Lyons (2021),we proposed a novel approach reshaping signature calculations to leverage GPU parallelism. This adjustment allows us to reduce the training time by 55\\% and 5 to 10-fold improvements in direct signature computation compared to existing methods, while maintaining"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2501.08455","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/2501.08455/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":"2501.08455","created_at":"2026-07-05T10:01:20.204712+00:00"},{"alias_kind":"arxiv_version","alias_value":"2501.08455v1","created_at":"2026-07-05T10:01:20.204712+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2501.08455","created_at":"2026-07-05T10:01:20.204712+00:00"},{"alias_kind":"pith_short_12","alias_value":"MIVZPVIPGPVX","created_at":"2026-07-05T10:01:20.204712+00:00"},{"alias_kind":"pith_short_16","alias_value":"MIVZPVIPGPVXALAG","created_at":"2026-07-05T10:01:20.204712+00:00"},{"alias_kind":"pith_short_8","alias_value":"MIVZPVIP","created_at":"2026-07-05T10:01:20.204712+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2502.09318","citing_title":"SigGate: Enhancing Recurrent Neural Networks with Signature-Based Gating Mechanisms","ref_index":25,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/MIVZPVIPGPVXALAGHPEIQPZXMO","json":"https://pith.science/pith/MIVZPVIPGPVXALAGHPEIQPZXMO.json","graph_json":"https://pith.science/api/pith-number/MIVZPVIPGPVXALAGHPEIQPZXMO/graph.json","events_json":"https://pith.science/api/pith-number/MIVZPVIPGPVXALAGHPEIQPZXMO/events.json","paper":"https://pith.science/paper/MIVZPVIP"},"agent_actions":{"view_html":"https://pith.science/pith/MIVZPVIPGPVXALAGHPEIQPZXMO","download_json":"https://pith.science/pith/MIVZPVIPGPVXALAGHPEIQPZXMO.json","view_paper":"https://pith.science/paper/MIVZPVIP","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2501.08455&json=true","fetch_graph":"https://pith.science/api/pith-number/MIVZPVIPGPVXALAGHPEIQPZXMO/graph.json","fetch_events":"https://pith.science/api/pith-number/MIVZPVIPGPVXALAGHPEIQPZXMO/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/MIVZPVIPGPVXALAGHPEIQPZXMO/action/timestamp_anchor","attest_storage":"https://pith.science/pith/MIVZPVIPGPVXALAGHPEIQPZXMO/action/storage_attestation","attest_author":"https://pith.science/pith/MIVZPVIPGPVXALAGHPEIQPZXMO/action/author_attestation","sign_citation":"https://pith.science/pith/MIVZPVIPGPVXALAGHPEIQPZXMO/action/citation_signature","submit_replication":"https://pith.science/pith/MIVZPVIPGPVXALAGHPEIQPZXMO/action/replication_record"}},"created_at":"2026-07-05T10:01:20.204712+00:00","updated_at":"2026-07-05T10:01:20.204712+00:00"}