{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:ZYT6CM6IRHF7DBNHZ3X3LL3RE2","short_pith_number":"pith:ZYT6CM6I","schema_version":"1.0","canonical_sha256":"ce27e133c889cbf185a7ceefb5af712680415bcf1f31e65f0d844617834e3b2d","source":{"kind":"arxiv","id":"2203.03366","version":1},"attestation_state":"computed","paper":{"title":"Improvements to Gradient Descent Methods for Quantum Tensor Network Machine Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.str-el","quant-ph"],"primary_cat":"cs.LG","authors_text":"Fergus Barratt, James Dborin, Lewis Wright","submitted_at":"2022-03-03T19:00:40Z","abstract_excerpt":"Tensor networks have demonstrated significant value for machine learning in a myriad of different applications. However, optimizing tensor networks using standard gradient descent has proven to be difficult in practice. Tensor networks suffer from initialization problems resulting in exploding or vanishing gradients and require extensive hyperparameter tuning. Efforts to overcome these problems usually depend on specific network architectures, or ad hoc prescriptions. In this paper we address the problems of initialization and hyperparameter tuning, making it possible to train tensor networks "},"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":"2203.03366","kind":"arxiv","version":1},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2022-03-03T19:00:40Z","cross_cats_sorted":["cond-mat.str-el","quant-ph"],"title_canon_sha256":"1e396597aaa798dd2d45a5fed65b2dc9e14c5933ab1599913cfcb477981a4c86","abstract_canon_sha256":"0c9b75e5064539bbf4add00af741f673c02e74d3e238a0793884cdb603005579"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T04:02:33.829210Z","signature_b64":"sHVjmSNJ85e5lODZJBheHYa3GieroB1l24VOVd29zLKBP3WzAw8ji+SaMgUl5vtwbq8jqT69438lAaXUMYGeDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ce27e133c889cbf185a7ceefb5af712680415bcf1f31e65f0d844617834e3b2d","last_reissued_at":"2026-07-05T04:02:33.828764Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T04:02:33.828764Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Improvements to Gradient Descent Methods for Quantum Tensor Network Machine Learning","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cond-mat.str-el","quant-ph"],"primary_cat":"cs.LG","authors_text":"Fergus Barratt, James Dborin, Lewis Wright","submitted_at":"2022-03-03T19:00:40Z","abstract_excerpt":"Tensor networks have demonstrated significant value for machine learning in a myriad of different applications. However, optimizing tensor networks using standard gradient descent has proven to be difficult in practice. Tensor networks suffer from initialization problems resulting in exploding or vanishing gradients and require extensive hyperparameter tuning. Efforts to overcome these problems usually depend on specific network architectures, or ad hoc prescriptions. In this paper we address the problems of initialization and hyperparameter tuning, making it possible to train tensor networks "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2203.03366","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/2203.03366/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":"2203.03366","created_at":"2026-07-05T04:02:33.828826+00:00"},{"alias_kind":"arxiv_version","alias_value":"2203.03366v1","created_at":"2026-07-05T04:02:33.828826+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2203.03366","created_at":"2026-07-05T04:02:33.828826+00:00"},{"alias_kind":"pith_short_12","alias_value":"ZYT6CM6IRHF7","created_at":"2026-07-05T04:02:33.828826+00:00"},{"alias_kind":"pith_short_16","alias_value":"ZYT6CM6IRHF7DBNH","created_at":"2026-07-05T04:02:33.828826+00:00"},{"alias_kind":"pith_short_8","alias_value":"ZYT6CM6I","created_at":"2026-07-05T04:02:33.828826+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2309.06577","citing_title":"Efficient Finite Initialization with Partial Norms for Tensorized Neural Networks and Tensor Networks Algorithms","ref_index":32,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/ZYT6CM6IRHF7DBNHZ3X3LL3RE2","json":"https://pith.science/pith/ZYT6CM6IRHF7DBNHZ3X3LL3RE2.json","graph_json":"https://pith.science/api/pith-number/ZYT6CM6IRHF7DBNHZ3X3LL3RE2/graph.json","events_json":"https://pith.science/api/pith-number/ZYT6CM6IRHF7DBNHZ3X3LL3RE2/events.json","paper":"https://pith.science/paper/ZYT6CM6I"},"agent_actions":{"view_html":"https://pith.science/pith/ZYT6CM6IRHF7DBNHZ3X3LL3RE2","download_json":"https://pith.science/pith/ZYT6CM6IRHF7DBNHZ3X3LL3RE2.json","view_paper":"https://pith.science/paper/ZYT6CM6I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2203.03366&json=true","fetch_graph":"https://pith.science/api/pith-number/ZYT6CM6IRHF7DBNHZ3X3LL3RE2/graph.json","fetch_events":"https://pith.science/api/pith-number/ZYT6CM6IRHF7DBNHZ3X3LL3RE2/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/ZYT6CM6IRHF7DBNHZ3X3LL3RE2/action/timestamp_anchor","attest_storage":"https://pith.science/pith/ZYT6CM6IRHF7DBNHZ3X3LL3RE2/action/storage_attestation","attest_author":"https://pith.science/pith/ZYT6CM6IRHF7DBNHZ3X3LL3RE2/action/author_attestation","sign_citation":"https://pith.science/pith/ZYT6CM6IRHF7DBNHZ3X3LL3RE2/action/citation_signature","submit_replication":"https://pith.science/pith/ZYT6CM6IRHF7DBNHZ3X3LL3RE2/action/replication_record"}},"created_at":"2026-07-05T04:02:33.828826+00:00","updated_at":"2026-07-05T04:02:33.828826+00:00"}