{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2019:6Z6ZTGV3CKNMUD4APNBGFXJCHQ","short_pith_number":"pith:6Z6ZTGV3","schema_version":"1.0","canonical_sha256":"f67d999abb129aca0f807b4262dd223c3dcfb1ad0ee33aae5e103f2bf85bc7a6","source":{"kind":"arxiv","id":"1902.01785","version":4},"attestation_state":"computed","paper":{"title":"Homogeneous Linear Inequality Constraints for Neural Network Activations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel Cremers, Matthias Nie{\\ss}ner, Thomas Frerix","submitted_at":"2019-02-05T16:44:55Z","abstract_excerpt":"We propose a method to impose homogeneous linear inequality constraints of the form $Ax\\leq 0$ on neural network activations. The proposed method allows a data-driven training approach to be combined with modeling prior knowledge about the task. One way to achieve this task is by means of a projection step at test time after unconstrained training. However, this is an expensive operation. By directly incorporating the constraints into the architecture, we can significantly speed-up inference at test time; for instance, our experiments show a speed-up of up to two orders of magnitude over a pro"},"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":"1902.01785","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-02-05T16:44:55Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"55fb670f5c825021a956efe97df1fdc963f9797b050a48ef75ee16b234d2aae9","abstract_canon_sha256":"35f6d1192367251719c4539d1f0c802ecdb62ff7eabe6777065d79ada8733390"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T01:06:17.514589Z","signature_b64":"2SYK9qwF2DoPHAHmZ7tPbXkRkQz5Q7si3FNmkYmxmTCdyrttsc91zwu8fsqZ0nDRGnad36VQL/oPB33iiIpKBA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f67d999abb129aca0f807b4262dd223c3dcfb1ad0ee33aae5e103f2bf85bc7a6","last_reissued_at":"2026-07-05T01:06:17.514029Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T01:06:17.514029Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Homogeneous Linear Inequality Constraints for Neural Network Activations","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Daniel Cremers, Matthias Nie{\\ss}ner, Thomas Frerix","submitted_at":"2019-02-05T16:44:55Z","abstract_excerpt":"We propose a method to impose homogeneous linear inequality constraints of the form $Ax\\leq 0$ on neural network activations. The proposed method allows a data-driven training approach to be combined with modeling prior knowledge about the task. One way to achieve this task is by means of a projection step at test time after unconstrained training. However, this is an expensive operation. By directly incorporating the constraints into the architecture, we can significantly speed-up inference at test time; for instance, our experiments show a speed-up of up to two orders of magnitude over a pro"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1902.01785","kind":"arxiv","version":4},"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/1902.01785/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":"1902.01785","created_at":"2026-07-05T01:06:17.514089+00:00"},{"alias_kind":"arxiv_version","alias_value":"1902.01785v4","created_at":"2026-07-05T01:06:17.514089+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1902.01785","created_at":"2026-07-05T01:06:17.514089+00:00"},{"alias_kind":"pith_short_12","alias_value":"6Z6ZTGV3CKNM","created_at":"2026-07-05T01:06:17.514089+00:00"},{"alias_kind":"pith_short_16","alias_value":"6Z6ZTGV3CKNMUD4A","created_at":"2026-07-05T01:06:17.514089+00:00"},{"alias_kind":"pith_short_8","alias_value":"6Z6ZTGV3","created_at":"2026-07-05T01:06:17.514089+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2602.09317","citing_title":"SnareNet: Flexible Repair Layers for Neural Networks with Hard Constraints","ref_index":12,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6Z6ZTGV3CKNMUD4APNBGFXJCHQ","json":"https://pith.science/pith/6Z6ZTGV3CKNMUD4APNBGFXJCHQ.json","graph_json":"https://pith.science/api/pith-number/6Z6ZTGV3CKNMUD4APNBGFXJCHQ/graph.json","events_json":"https://pith.science/api/pith-number/6Z6ZTGV3CKNMUD4APNBGFXJCHQ/events.json","paper":"https://pith.science/paper/6Z6ZTGV3"},"agent_actions":{"view_html":"https://pith.science/pith/6Z6ZTGV3CKNMUD4APNBGFXJCHQ","download_json":"https://pith.science/pith/6Z6ZTGV3CKNMUD4APNBGFXJCHQ.json","view_paper":"https://pith.science/paper/6Z6ZTGV3","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1902.01785&json=true","fetch_graph":"https://pith.science/api/pith-number/6Z6ZTGV3CKNMUD4APNBGFXJCHQ/graph.json","fetch_events":"https://pith.science/api/pith-number/6Z6ZTGV3CKNMUD4APNBGFXJCHQ/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6Z6ZTGV3CKNMUD4APNBGFXJCHQ/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6Z6ZTGV3CKNMUD4APNBGFXJCHQ/action/storage_attestation","attest_author":"https://pith.science/pith/6Z6ZTGV3CKNMUD4APNBGFXJCHQ/action/author_attestation","sign_citation":"https://pith.science/pith/6Z6ZTGV3CKNMUD4APNBGFXJCHQ/action/citation_signature","submit_replication":"https://pith.science/pith/6Z6ZTGV3CKNMUD4APNBGFXJCHQ/action/replication_record"}},"created_at":"2026-07-05T01:06:17.514089+00:00","updated_at":"2026-07-05T01:06:17.514089+00:00"}