{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:CYEU32C2Y5A6247EMVMPF7QB3S","short_pith_number":"pith:CYEU32C2","schema_version":"1.0","canonical_sha256":"16094de85ac741ed73e46558f2fe01dcbb7b9bc9c81d779205d93b356df94028","source":{"kind":"arxiv","id":"2105.01957","version":1},"attestation_state":"computed","paper":{"title":"Perceptual Gradient Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Aleksei Ivakhnenko, Dmitry Nikulin, Roman Suvorov, Victor Lempitsky","submitted_at":"2021-05-05T09:58:22Z","abstract_excerpt":"Many applications of deep learning for image generation use perceptual losses for either training or fine-tuning of the generator networks. The use of perceptual loss however incurs repeated forward-backward passes in a large image classification network as well as a considerable memory overhead required to store the activations of this network. It is therefore desirable or sometimes even critical to get rid of these overheads.\n  In this work, we propose a way to train generator networks using approximations of perceptual loss that are computed without forward-backward passes. Instead, we use "},"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":"2105.01957","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-05-05T09:58:22Z","cross_cats_sorted":["cs.CV"],"title_canon_sha256":"36234021232fcbeabfe909c3bfeee89f27f46d58ca82a9545aac03f5fab1f032","abstract_canon_sha256":"5aded8999b461f46958629b0b040e4c423759d406f8a38b23d79830123240316"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:37:50.832836Z","signature_b64":"8eeNdeB1665Ru0NlbUDADais7xojDybG+5pVUKhsoPMWkC+7Sjd5nPoBM1A7ABJx4ZikXBRaOQbLLzDLjwgwCQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"16094de85ac741ed73e46558f2fe01dcbb7b9bc9c81d779205d93b356df94028","last_reissued_at":"2026-07-05T02:37:50.832403Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:37:50.832403Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Perceptual Gradient Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.CV"],"primary_cat":"cs.LG","authors_text":"Aleksei Ivakhnenko, Dmitry Nikulin, Roman Suvorov, Victor Lempitsky","submitted_at":"2021-05-05T09:58:22Z","abstract_excerpt":"Many applications of deep learning for image generation use perceptual losses for either training or fine-tuning of the generator networks. The use of perceptual loss however incurs repeated forward-backward passes in a large image classification network as well as a considerable memory overhead required to store the activations of this network. It is therefore desirable or sometimes even critical to get rid of these overheads.\n  In this work, we propose a way to train generator networks using approximations of perceptual loss that are computed without forward-backward passes. Instead, we use "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2105.01957","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/2105.01957/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":"2105.01957","created_at":"2026-07-05T02:37:50.832462+00:00"},{"alias_kind":"arxiv_version","alias_value":"2105.01957v1","created_at":"2026-07-05T02:37:50.832462+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2105.01957","created_at":"2026-07-05T02:37:50.832462+00:00"},{"alias_kind":"pith_short_12","alias_value":"CYEU32C2Y5A6","created_at":"2026-07-05T02:37:50.832462+00:00"},{"alias_kind":"pith_short_16","alias_value":"CYEU32C2Y5A6247E","created_at":"2026-07-05T02:37:50.832462+00:00"},{"alias_kind":"pith_short_8","alias_value":"CYEU32C2","created_at":"2026-07-05T02:37:50.832462+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/CYEU32C2Y5A6247EMVMPF7QB3S","json":"https://pith.science/pith/CYEU32C2Y5A6247EMVMPF7QB3S.json","graph_json":"https://pith.science/api/pith-number/CYEU32C2Y5A6247EMVMPF7QB3S/graph.json","events_json":"https://pith.science/api/pith-number/CYEU32C2Y5A6247EMVMPF7QB3S/events.json","paper":"https://pith.science/paper/CYEU32C2"},"agent_actions":{"view_html":"https://pith.science/pith/CYEU32C2Y5A6247EMVMPF7QB3S","download_json":"https://pith.science/pith/CYEU32C2Y5A6247EMVMPF7QB3S.json","view_paper":"https://pith.science/paper/CYEU32C2","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2105.01957&json=true","fetch_graph":"https://pith.science/api/pith-number/CYEU32C2Y5A6247EMVMPF7QB3S/graph.json","fetch_events":"https://pith.science/api/pith-number/CYEU32C2Y5A6247EMVMPF7QB3S/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/CYEU32C2Y5A6247EMVMPF7QB3S/action/timestamp_anchor","attest_storage":"https://pith.science/pith/CYEU32C2Y5A6247EMVMPF7QB3S/action/storage_attestation","attest_author":"https://pith.science/pith/CYEU32C2Y5A6247EMVMPF7QB3S/action/author_attestation","sign_citation":"https://pith.science/pith/CYEU32C2Y5A6247EMVMPF7QB3S/action/citation_signature","submit_replication":"https://pith.science/pith/CYEU32C2Y5A6247EMVMPF7QB3S/action/replication_record"}},"created_at":"2026-07-05T02:37:50.832462+00:00","updated_at":"2026-07-05T02:37:50.832462+00:00"}