{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2022:CITUMKLLD6B3G3VQAIGC2XLG7O","short_pith_number":"pith:CITUMKLL","schema_version":"1.0","canonical_sha256":"122746296b1f83b36eb0020c2d5d66fb874d7ca29320bc9ddff6ebfa2e9615a2","source":{"kind":"arxiv","id":"2209.13435","version":2},"attestation_state":"computed","paper":{"title":"Scaling Laws For Deep Learning Based Image Reconstruction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG","stat.ML"],"primary_cat":"eess.IV","authors_text":"Reinhard Heckel, Tobit Klug","submitted_at":"2022-09-27T14:44:57Z","abstract_excerpt":"Deep neural networks trained end-to-end to map a measurement of a (noisy) image to a clean image perform excellent for a variety of linear inverse problems. Current methods are only trained on a few hundreds or thousands of images as opposed to the millions of examples deep networks are trained on in other domains. In this work, we study whether major performance gains are expected from scaling up the training set size. We consider image denoising, accelerated magnetic resonance imaging, and super-resolution and empirically determine the reconstruction quality as a function of training set siz"},"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":"2209.13435","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"eess.IV","submitted_at":"2022-09-27T14:44:57Z","cross_cats_sorted":["cs.AI","cs.CV","cs.LG","stat.ML"],"title_canon_sha256":"894d258186063b2f9e1a71f5f3b939b866f4c266c907e7d1a3e163370ce2c925","abstract_canon_sha256":"2a1b7c23ca0bd45ff28096d6c13a7853c036eef884601c85470aa794e5334041"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:44:50.639386Z","signature_b64":"w5rsGzXQ6axX6X0jqZWHcRCYHevln1y8ixlKVn5nRoxlANK5ciZ7v1H72johcKwG6Ywa+Hl0leQ1QMgO+WfnCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"122746296b1f83b36eb0020c2d5d66fb874d7ca29320bc9ddff6ebfa2e9615a2","last_reissued_at":"2026-07-05T05:44:50.638847Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:44:50.638847Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Scaling Laws For Deep Learning Based Image Reconstruction","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","cs.CV","cs.LG","stat.ML"],"primary_cat":"eess.IV","authors_text":"Reinhard Heckel, Tobit Klug","submitted_at":"2022-09-27T14:44:57Z","abstract_excerpt":"Deep neural networks trained end-to-end to map a measurement of a (noisy) image to a clean image perform excellent for a variety of linear inverse problems. Current methods are only trained on a few hundreds or thousands of images as opposed to the millions of examples deep networks are trained on in other domains. In this work, we study whether major performance gains are expected from scaling up the training set size. We consider image denoising, accelerated magnetic resonance imaging, and super-resolution and empirically determine the reconstruction quality as a function of training set siz"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2209.13435","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/2209.13435/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":"2209.13435","created_at":"2026-07-05T05:44:50.638910+00:00"},{"alias_kind":"arxiv_version","alias_value":"2209.13435v2","created_at":"2026-07-05T05:44:50.638910+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2209.13435","created_at":"2026-07-05T05:44:50.638910+00:00"},{"alias_kind":"pith_short_12","alias_value":"CITUMKLLD6B3","created_at":"2026-07-05T05:44:50.638910+00:00"},{"alias_kind":"pith_short_16","alias_value":"CITUMKLLD6B3G3VQ","created_at":"2026-07-05T05:44:50.638910+00:00"},{"alias_kind":"pith_short_8","alias_value":"CITUMKLL","created_at":"2026-07-05T05:44:50.638910+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2606.06725","citing_title":"Compute-Optimal Network Design for Echocardiography Myocardial Segmentation and Perfusion Quantification using Neural Scaling Laws","ref_index":17,"is_internal_anchor":false},{"citing_arxiv_id":"2509.24244","citing_title":"Model Merging Scaling Laws in Large Language Models","ref_index":13,"is_internal_anchor":false},{"citing_arxiv_id":"2211.01324","citing_title":"eDiff-I: Text-to-Image Diffusion Models with an Ensemble of Expert Denoisers","ref_index":38,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/CITUMKLLD6B3G3VQAIGC2XLG7O","json":"https://pith.science/pith/CITUMKLLD6B3G3VQAIGC2XLG7O.json","graph_json":"https://pith.science/api/pith-number/CITUMKLLD6B3G3VQAIGC2XLG7O/graph.json","events_json":"https://pith.science/api/pith-number/CITUMKLLD6B3G3VQAIGC2XLG7O/events.json","paper":"https://pith.science/paper/CITUMKLL"},"agent_actions":{"view_html":"https://pith.science/pith/CITUMKLLD6B3G3VQAIGC2XLG7O","download_json":"https://pith.science/pith/CITUMKLLD6B3G3VQAIGC2XLG7O.json","view_paper":"https://pith.science/paper/CITUMKLL","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2209.13435&json=true","fetch_graph":"https://pith.science/api/pith-number/CITUMKLLD6B3G3VQAIGC2XLG7O/graph.json","fetch_events":"https://pith.science/api/pith-number/CITUMKLLD6B3G3VQAIGC2XLG7O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CITUMKLLD6B3G3VQAIGC2XLG7O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CITUMKLLD6B3G3VQAIGC2XLG7O/action/storage_attestation","attest_author":"https://pith.science/pith/CITUMKLLD6B3G3VQAIGC2XLG7O/action/author_attestation","sign_citation":"https://pith.science/pith/CITUMKLLD6B3G3VQAIGC2XLG7O/action/citation_signature","submit_replication":"https://pith.science/pith/CITUMKLLD6B3G3VQAIGC2XLG7O/action/replication_record"}},"created_at":"2026-07-05T05:44:50.638910+00:00","updated_at":"2026-07-05T05:44:50.638910+00:00"}