{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:DNUB75EXPRS35U5LJDRQPU3J6E","short_pith_number":"pith:DNUB75EX","schema_version":"1.0","canonical_sha256":"1b681ff4977c65bed3ab48e307d369f125ca97e04696c0cd00521d60d8c38622","source":{"kind":"arxiv","id":"1908.09998","version":1},"attestation_state":"computed","paper":{"title":"Distorted Representation Space Characterization Through Backpropagated Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.IV"],"primary_cat":"cs.CV","authors_text":"Dogancan Temel, Ghassan AlRegib, Gukyeong Kwon, Mohit Prabhushankar","submitted_at":"2019-08-27T02:58:43Z","abstract_excerpt":"In this paper, we utilize weight gradients from backpropagation to characterize the representation space learned by deep learning algorithms. We demonstrate the utility of such gradients in applications including perceptual image quality assessment and out-of-distribution classification. The applications are chosen to validate the effectiveness of gradients as features when the test image distribution is distorted from the train image distribution. In both applications, the proposed gradient based features outperform activation features. In image quality assessment, the proposed approach is co"},"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":"1908.09998","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2019-08-27T02:58:43Z","cross_cats_sorted":["eess.IV"],"title_canon_sha256":"b3a0109b0df0090cf9ad15d4733439f8ffbcdf6ec1f2bec2e9bda09a9594918f","abstract_canon_sha256":"99e8cf796e781c7a824ef9500eca50a0d0b6c36ddf53a27e8229f0f53e561c03"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-04T23:59:52.725083Z","signature_b64":"O2kixUnI7pkKq5i/UxoKf+FhRXCXcsNGs9flYRgQ5nsWZYPw9C44ZEynVchBNPFhBHrcs+9XFc3wrTa3a8oIDA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"1b681ff4977c65bed3ab48e307d369f125ca97e04696c0cd00521d60d8c38622","last_reissued_at":"2026-07-04T23:59:52.724609Z","signature_status":"signed_v1","first_computed_at":"2026-07-04T23:59:52.724609Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Distorted Representation Space Characterization Through Backpropagated Gradients","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["eess.IV"],"primary_cat":"cs.CV","authors_text":"Dogancan Temel, Ghassan AlRegib, Gukyeong Kwon, Mohit Prabhushankar","submitted_at":"2019-08-27T02:58:43Z","abstract_excerpt":"In this paper, we utilize weight gradients from backpropagation to characterize the representation space learned by deep learning algorithms. We demonstrate the utility of such gradients in applications including perceptual image quality assessment and out-of-distribution classification. The applications are chosen to validate the effectiveness of gradients as features when the test image distribution is distorted from the train image distribution. In both applications, the proposed gradient based features outperform activation features. In image quality assessment, the proposed approach is co"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1908.09998","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/1908.09998/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":"1908.09998","created_at":"2026-07-04T23:59:52.724660+00:00"},{"alias_kind":"arxiv_version","alias_value":"1908.09998v1","created_at":"2026-07-04T23:59:52.724660+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1908.09998","created_at":"2026-07-04T23:59:52.724660+00:00"},{"alias_kind":"pith_short_12","alias_value":"DNUB75EXPRS3","created_at":"2026-07-04T23:59:52.724660+00:00"},{"alias_kind":"pith_short_16","alias_value":"DNUB75EXPRS35U5L","created_at":"2026-07-04T23:59:52.724660+00:00"},{"alias_kind":"pith_short_8","alias_value":"DNUB75EX","created_at":"2026-07-04T23:59:52.724660+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/DNUB75EXPRS35U5LJDRQPU3J6E","json":"https://pith.science/pith/DNUB75EXPRS35U5LJDRQPU3J6E.json","graph_json":"https://pith.science/api/pith-number/DNUB75EXPRS35U5LJDRQPU3J6E/graph.json","events_json":"https://pith.science/api/pith-number/DNUB75EXPRS35U5LJDRQPU3J6E/events.json","paper":"https://pith.science/paper/DNUB75EX"},"agent_actions":{"view_html":"https://pith.science/pith/DNUB75EXPRS35U5LJDRQPU3J6E","download_json":"https://pith.science/pith/DNUB75EXPRS35U5LJDRQPU3J6E.json","view_paper":"https://pith.science/paper/DNUB75EX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1908.09998&json=true","fetch_graph":"https://pith.science/api/pith-number/DNUB75EXPRS35U5LJDRQPU3J6E/graph.json","fetch_events":"https://pith.science/api/pith-number/DNUB75EXPRS35U5LJDRQPU3J6E/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/DNUB75EXPRS35U5LJDRQPU3J6E/action/timestamp_anchor","attest_storage":"https://pith.science/pith/DNUB75EXPRS35U5LJDRQPU3J6E/action/storage_attestation","attest_author":"https://pith.science/pith/DNUB75EXPRS35U5LJDRQPU3J6E/action/author_attestation","sign_citation":"https://pith.science/pith/DNUB75EXPRS35U5LJDRQPU3J6E/action/citation_signature","submit_replication":"https://pith.science/pith/DNUB75EXPRS35U5LJDRQPU3J6E/action/replication_record"}},"created_at":"2026-07-04T23:59:52.724660+00:00","updated_at":"2026-07-04T23:59:52.724660+00:00"}