{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HLZYUBTDUSPJXA7FNU6FSWR663","short_pith_number":"pith:HLZYUBTD","schema_version":"1.0","canonical_sha256":"3af38a0663a49e9b83e56d3c595a3ef6df499f36d46da3b370dcd10371daf012","source":{"kind":"arxiv","id":"2103.14616","version":2},"attestation_state":"computed","paper":{"title":"Training a Task-Specific Image Reconstruction Loss","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"eess.IV","authors_text":"Aamir Mustafa, Aliaksei Mikhailiuk, Dan Andrei Iliescu, Rafal K. Mantiuk, Varun Babbar","submitted_at":"2021-03-26T17:29:57Z","abstract_excerpt":"The choice of a loss function is an important factor when training neural networks for image restoration problems, such as single image super resolution. The loss function should encourage natural and perceptually pleasing results. A popular choice for a loss is a pre-trained network, such as VGG, which is used as a feature extractor for computing the difference between restored and reference images. However, such an approach has multiple drawbacks: it is computationally expensive, requires regularization and hyper-parameter tuning, and involves a large network trained on an unrelated task. Fu"},"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":"2103.14616","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"eess.IV","submitted_at":"2021-03-26T17:29:57Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"94c39049fda7161874a2ab31ea2f4710f30055cbe427c49391d5f92b74824ce6","abstract_canon_sha256":"99919500b5248a5f63d94c1e740371763787004e100d2d088464d04d812f7460"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:23:08.786499Z","signature_b64":"3/TU8WpveA71FeSuf59rB1APPJ9RSEqriLNDTMaiJxykTyv/9B7y0acjmNUxnkWB2XIeWrjYyCL1FmIVq0HJCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3af38a0663a49e9b83e56d3c595a3ef6df499f36d46da3b370dcd10371daf012","last_reissued_at":"2026-07-05T03:23:08.786046Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:23:08.786046Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Training a Task-Specific Image Reconstruction Loss","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"eess.IV","authors_text":"Aamir Mustafa, Aliaksei Mikhailiuk, Dan Andrei Iliescu, Rafal K. Mantiuk, Varun Babbar","submitted_at":"2021-03-26T17:29:57Z","abstract_excerpt":"The choice of a loss function is an important factor when training neural networks for image restoration problems, such as single image super resolution. The loss function should encourage natural and perceptually pleasing results. A popular choice for a loss is a pre-trained network, such as VGG, which is used as a feature extractor for computing the difference between restored and reference images. However, such an approach has multiple drawbacks: it is computationally expensive, requires regularization and hyper-parameter tuning, and involves a large network trained on an unrelated task. Fu"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.14616","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/2103.14616/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":"2103.14616","created_at":"2026-07-05T03:23:08.786118+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.14616v2","created_at":"2026-07-05T03:23:08.786118+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.14616","created_at":"2026-07-05T03:23:08.786118+00:00"},{"alias_kind":"pith_short_12","alias_value":"HLZYUBTDUSPJ","created_at":"2026-07-05T03:23:08.786118+00:00"},{"alias_kind":"pith_short_16","alias_value":"HLZYUBTDUSPJXA7F","created_at":"2026-07-05T03:23:08.786118+00:00"},{"alias_kind":"pith_short_8","alias_value":"HLZYUBTD","created_at":"2026-07-05T03:23:08.786118+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":0,"internal_anchor_count":0,"sample":[]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HLZYUBTDUSPJXA7FNU6FSWR663","json":"https://pith.science/pith/HLZYUBTDUSPJXA7FNU6FSWR663.json","graph_json":"https://pith.science/api/pith-number/HLZYUBTDUSPJXA7FNU6FSWR663/graph.json","events_json":"https://pith.science/api/pith-number/HLZYUBTDUSPJXA7FNU6FSWR663/events.json","paper":"https://pith.science/paper/HLZYUBTD"},"agent_actions":{"view_html":"https://pith.science/pith/HLZYUBTDUSPJXA7FNU6FSWR663","download_json":"https://pith.science/pith/HLZYUBTDUSPJXA7FNU6FSWR663.json","view_paper":"https://pith.science/paper/HLZYUBTD","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.14616&json=true","fetch_graph":"https://pith.science/api/pith-number/HLZYUBTDUSPJXA7FNU6FSWR663/graph.json","fetch_events":"https://pith.science/api/pith-number/HLZYUBTDUSPJXA7FNU6FSWR663/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HLZYUBTDUSPJXA7FNU6FSWR663/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HLZYUBTDUSPJXA7FNU6FSWR663/action/storage_attestation","attest_author":"https://pith.science/pith/HLZYUBTDUSPJXA7FNU6FSWR663/action/author_attestation","sign_citation":"https://pith.science/pith/HLZYUBTDUSPJXA7FNU6FSWR663/action/citation_signature","submit_replication":"https://pith.science/pith/HLZYUBTDUSPJXA7FNU6FSWR663/action/replication_record"}},"created_at":"2026-07-05T03:23:08.786118+00:00","updated_at":"2026-07-05T03:23:08.786118+00:00"}