{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:C6ZALRCQ323UR532J75QVFDEFU","short_pith_number":"pith:C6ZALRCQ","schema_version":"1.0","canonical_sha256":"17b205c450deb748f77a4ffb0a94642d2a256314db084bf10e4ed58e7ecf9201","source":{"kind":"arxiv","id":"2301.11360","version":2},"attestation_state":"computed","paper":{"title":"The Power of Linear Combinations: Learning with Random Convolutions","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Janis Keuper, Paul Gavrikov","submitted_at":"2023-01-26T19:17:10Z","abstract_excerpt":"Following the traditional paradigm of convolutional neural networks (CNNs), modern CNNs manage to keep pace with more recent, for example transformer-based, models by not only increasing model depth and width but also the kernel size. This results in large amounts of learnable model parameters that need to be handled during training. While following the convolutional paradigm with the according spatial inductive bias, we question the significance of \\emph{learned} convolution filters. In fact, our findings demonstrate that many contemporary CNN architectures can achieve high test accuracies wi"},"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":"2301.11360","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by-sa/4.0/","primary_cat":"cs.CV","submitted_at":"2023-01-26T19:17:10Z","cross_cats_sorted":["cs.AI","cs.LG"],"title_canon_sha256":"7d7ce4c1fbe718d9a90ba7958d74d0505a5251f68ecb91c7bab1b75421b02b98","abstract_canon_sha256":"07241335356ca9d680a02b8aa4a918d305a2d42ffdb7c59647c0f325e4e0c378"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T06:23:37.562852Z","signature_b64":"b+E4pBKNsGw1vJUun2e/cAi6WRcqrL7L3sx6ACAnsX0Hy7EPXuKuIcpjeJrnSzl7Rt9Vr+9wdlIfqIpgfwCGDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"17b205c450deb748f77a4ffb0a94642d2a256314db084bf10e4ed58e7ecf9201","last_reissued_at":"2026-07-05T06:23:37.562388Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T06:23:37.562388Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Power of Linear Combinations: Learning with Random Convolutions","license":"http://creativecommons.org/licenses/by-sa/4.0/","headline":"","cross_cats":["cs.AI","cs.LG"],"primary_cat":"cs.CV","authors_text":"Janis Keuper, Paul Gavrikov","submitted_at":"2023-01-26T19:17:10Z","abstract_excerpt":"Following the traditional paradigm of convolutional neural networks (CNNs), modern CNNs manage to keep pace with more recent, for example transformer-based, models by not only increasing model depth and width but also the kernel size. This results in large amounts of learnable model parameters that need to be handled during training. While following the convolutional paradigm with the according spatial inductive bias, we question the significance of \\emph{learned} convolution filters. In fact, our findings demonstrate that many contemporary CNN architectures can achieve high test accuracies wi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2301.11360","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/2301.11360/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":"2301.11360","created_at":"2026-07-05T06:23:37.562443+00:00"},{"alias_kind":"arxiv_version","alias_value":"2301.11360v2","created_at":"2026-07-05T06:23:37.562443+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2301.11360","created_at":"2026-07-05T06:23:37.562443+00:00"},{"alias_kind":"pith_short_12","alias_value":"C6ZALRCQ323U","created_at":"2026-07-05T06:23:37.562443+00:00"},{"alias_kind":"pith_short_16","alias_value":"C6ZALRCQ323UR532","created_at":"2026-07-05T06:23:37.562443+00:00"},{"alias_kind":"pith_short_8","alias_value":"C6ZALRCQ","created_at":"2026-07-05T06:23:37.562443+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.12368","citing_title":"MetaColloc: Optimization-Free PDE Solving via Meta-Learned Basis Functions","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/C6ZALRCQ323UR532J75QVFDEFU","json":"https://pith.science/pith/C6ZALRCQ323UR532J75QVFDEFU.json","graph_json":"https://pith.science/api/pith-number/C6ZALRCQ323UR532J75QVFDEFU/graph.json","events_json":"https://pith.science/api/pith-number/C6ZALRCQ323UR532J75QVFDEFU/events.json","paper":"https://pith.science/paper/C6ZALRCQ"},"agent_actions":{"view_html":"https://pith.science/pith/C6ZALRCQ323UR532J75QVFDEFU","download_json":"https://pith.science/pith/C6ZALRCQ323UR532J75QVFDEFU.json","view_paper":"https://pith.science/paper/C6ZALRCQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2301.11360&json=true","fetch_graph":"https://pith.science/api/pith-number/C6ZALRCQ323UR532J75QVFDEFU/graph.json","fetch_events":"https://pith.science/api/pith-number/C6ZALRCQ323UR532J75QVFDEFU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/C6ZALRCQ323UR532J75QVFDEFU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/C6ZALRCQ323UR532J75QVFDEFU/action/storage_attestation","attest_author":"https://pith.science/pith/C6ZALRCQ323UR532J75QVFDEFU/action/author_attestation","sign_citation":"https://pith.science/pith/C6ZALRCQ323UR532J75QVFDEFU/action/citation_signature","submit_replication":"https://pith.science/pith/C6ZALRCQ323UR532J75QVFDEFU/action/replication_record"}},"created_at":"2026-07-05T06:23:37.562443+00:00","updated_at":"2026-07-05T06:23:37.562443+00:00"}