{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2025:6D5LXQU7OPD33Y4KLSP2XYBKN3","short_pith_number":"pith:6D5LXQU7","schema_version":"1.0","canonical_sha256":"f0fabbc29f73c7bde38a5c9fabe02a6ecc742e61f4374a20ca867d4530e32425","source":{"kind":"arxiv","id":"2505.11692","version":2},"attestation_state":"computed","paper":{"title":"The Geometry of ReLU Networks through the ReLU Transition Graph","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Sahil Rajesh Dhayalkar","submitted_at":"2025-05-16T21:00:56Z","abstract_excerpt":"We develop a novel theoretical framework for analyzing ReLU neural networks through the lens of a combinatorial object we term the ReLU Transition Graph (RTG). In this graph, each node corresponds to a linear region induced by the network's activation patterns, and edges connect regions that differ by a single neuron flip. Building on this structure, we derive a suite of new theoretical results connecting RTG geometry to expressivity, generalization, and robustness. Our contributions include tight combinatorial bounds on RTG size and diameter, a proof of RTG connectivity, and graph-theoretic i"},"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":"2505.11692","kind":"arxiv","version":2},"metadata":{"license":"http://creativecommons.org/licenses/by/4.0/","primary_cat":"cs.LG","submitted_at":"2025-05-16T21:00:56Z","cross_cats_sorted":["cs.AI","cs.NE"],"title_canon_sha256":"a17ccd30c9b1dc5114f83fe8d0431909fe4dfd2cc6477dd86dfa003e45f07a17","abstract_canon_sha256":"f80471b40e991b1fc4bf88b350689d00a023ebbf8bf3d1e4a7a88c9b87b100ac"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T11:11:45.447477Z","signature_b64":"ngwUF6cvPPwpPSlIjQ31H5kN2J4zlvUkgbNZgIkq+wjprujZ3pASczVRBgmtKWOj7ktowdLNrSVdo0M77CqPCg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f0fabbc29f73c7bde38a5c9fabe02a6ecc742e61f4374a20ca867d4530e32425","last_reissued_at":"2026-07-05T11:11:45.446988Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T11:11:45.446988Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"The Geometry of ReLU Networks through the ReLU Transition Graph","license":"http://creativecommons.org/licenses/by/4.0/","headline":"","cross_cats":["cs.AI","cs.NE"],"primary_cat":"cs.LG","authors_text":"Sahil Rajesh Dhayalkar","submitted_at":"2025-05-16T21:00:56Z","abstract_excerpt":"We develop a novel theoretical framework for analyzing ReLU neural networks through the lens of a combinatorial object we term the ReLU Transition Graph (RTG). In this graph, each node corresponds to a linear region induced by the network's activation patterns, and edges connect regions that differ by a single neuron flip. Building on this structure, we derive a suite of new theoretical results connecting RTG geometry to expressivity, generalization, and robustness. Our contributions include tight combinatorial bounds on RTG size and diameter, a proof of RTG connectivity, and graph-theoretic i"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2505.11692","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/2505.11692/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":"2505.11692","created_at":"2026-07-05T11:11:45.447042+00:00"},{"alias_kind":"arxiv_version","alias_value":"2505.11692v2","created_at":"2026-07-05T11:11:45.447042+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2505.11692","created_at":"2026-07-05T11:11:45.447042+00:00"},{"alias_kind":"pith_short_12","alias_value":"6D5LXQU7OPD3","created_at":"2026-07-05T11:11:45.447042+00:00"},{"alias_kind":"pith_short_16","alias_value":"6D5LXQU7OPD33Y4K","created_at":"2026-07-05T11:11:45.447042+00:00"},{"alias_kind":"pith_short_8","alias_value":"6D5LXQU7","created_at":"2026-07-05T11:11:45.447042+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2509.03056","citing_title":"Discrete Functional Geometry of ReLU Networks via ReLU Transition Graphs","ref_index":5,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/6D5LXQU7OPD33Y4KLSP2XYBKN3","json":"https://pith.science/pith/6D5LXQU7OPD33Y4KLSP2XYBKN3.json","graph_json":"https://pith.science/api/pith-number/6D5LXQU7OPD33Y4KLSP2XYBKN3/graph.json","events_json":"https://pith.science/api/pith-number/6D5LXQU7OPD33Y4KLSP2XYBKN3/events.json","paper":"https://pith.science/paper/6D5LXQU7"},"agent_actions":{"view_html":"https://pith.science/pith/6D5LXQU7OPD33Y4KLSP2XYBKN3","download_json":"https://pith.science/pith/6D5LXQU7OPD33Y4KLSP2XYBKN3.json","view_paper":"https://pith.science/paper/6D5LXQU7","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2505.11692&json=true","fetch_graph":"https://pith.science/api/pith-number/6D5LXQU7OPD33Y4KLSP2XYBKN3/graph.json","fetch_events":"https://pith.science/api/pith-number/6D5LXQU7OPD33Y4KLSP2XYBKN3/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6D5LXQU7OPD33Y4KLSP2XYBKN3/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6D5LXQU7OPD33Y4KLSP2XYBKN3/action/storage_attestation","attest_author":"https://pith.science/pith/6D5LXQU7OPD33Y4KLSP2XYBKN3/action/author_attestation","sign_citation":"https://pith.science/pith/6D5LXQU7OPD33Y4KLSP2XYBKN3/action/citation_signature","submit_replication":"https://pith.science/pith/6D5LXQU7OPD33Y4KLSP2XYBKN3/action/replication_record"}},"created_at":"2026-07-05T11:11:45.447042+00:00","updated_at":"2026-07-05T11:11:45.447042+00:00"}