{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2015:K53GI27IDFN4JDN7JNDC3PAHCH","short_pith_number":"pith:K53GI27I","schema_version":"1.0","canonical_sha256":"5776646be8195bc48dbf4b462dbc0711fdcc2373ac4e0505ef3580fb961e1b58","source":{"kind":"arxiv","id":"1511.06314","version":1},"attestation_state":"computed","paper":{"title":"Why M Heads are Better than One: Training a Diverse Ensemble of Deep Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"cs.CV","authors_text":"David Crandall, Dhruv Batra, Michael Cogswell, Senthil Purushwalkam, Stefan Lee","submitted_at":"2015-11-19T19:19:58Z","abstract_excerpt":"Convolutional Neural Networks have achieved state-of-the-art performance on a wide range of tasks. Most benchmarks are led by ensembles of these powerful learners, but ensembling is typically treated as a post-hoc procedure implemented by averaging independently trained models with model variation induced by bagging or random initialization. In this paper, we rigorously treat ensembling as a first-class problem to explicitly address the question: what are the best strategies to create an ensemble? We first compare a large number of ensembling strategies, and then propose and evaluate novel str"},"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":"1511.06314","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2015-11-19T19:19:58Z","cross_cats_sorted":["cs.LG","cs.NE"],"title_canon_sha256":"feea8ae58c9e10817e38c4836c0be948a95a2e96feb17b33fa4847b874c8bcaf","abstract_canon_sha256":"5dc5600768215955abac796f0341408f6cc23bfe57bf339e21f8721f66747faf"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T01:26:25.924755Z","signature_b64":"LWdVyj3cwLGRjuFT3pyBZRgI1DS/Yc+N/0dKkWAf+yjLku7k3bE4puyeS4yjGfOzh2sOovyEdNmcQUS1DHRWAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"5776646be8195bc48dbf4b462dbc0711fdcc2373ac4e0505ef3580fb961e1b58","last_reissued_at":"2026-05-18T01:26:25.924218Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T01:26:25.924218Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Why M Heads are Better than One: Training a Diverse Ensemble of Deep Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","cs.NE"],"primary_cat":"cs.CV","authors_text":"David Crandall, Dhruv Batra, Michael Cogswell, Senthil Purushwalkam, Stefan Lee","submitted_at":"2015-11-19T19:19:58Z","abstract_excerpt":"Convolutional Neural Networks have achieved state-of-the-art performance on a wide range of tasks. Most benchmarks are led by ensembles of these powerful learners, but ensembling is typically treated as a post-hoc procedure implemented by averaging independently trained models with model variation induced by bagging or random initialization. In this paper, we rigorously treat ensembling as a first-class problem to explicitly address the question: what are the best strategies to create an ensemble? We first compare a large number of ensembling strategies, and then propose and evaluate novel str"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1511.06314","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":""},"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":"1511.06314","created_at":"2026-05-18T01:26:25.924298+00:00"},{"alias_kind":"arxiv_version","alias_value":"1511.06314v1","created_at":"2026-05-18T01:26:25.924298+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1511.06314","created_at":"2026-05-18T01:26:25.924298+00:00"},{"alias_kind":"pith_short_12","alias_value":"K53GI27IDFN4","created_at":"2026-05-18T12:29:27.538025+00:00"},{"alias_kind":"pith_short_16","alias_value":"K53GI27IDFN4JDN7","created_at":"2026-05-18T12:29:27.538025+00:00"},{"alias_kind":"pith_short_8","alias_value":"K53GI27I","created_at":"2026-05-18T12:29:27.538025+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":2,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1907.11555","citing_title":"As easy as 1, 2... 4? Uncertainty in counting tasks for medical imaging","ref_index":3,"is_internal_anchor":true},{"citing_arxiv_id":"2605.04231","citing_title":"Anatomy of a failure: When, how, and why deep vision fails in scientific domains","ref_index":115,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/K53GI27IDFN4JDN7JNDC3PAHCH","json":"https://pith.science/pith/K53GI27IDFN4JDN7JNDC3PAHCH.json","graph_json":"https://pith.science/api/pith-number/K53GI27IDFN4JDN7JNDC3PAHCH/graph.json","events_json":"https://pith.science/api/pith-number/K53GI27IDFN4JDN7JNDC3PAHCH/events.json","paper":"https://pith.science/paper/K53GI27I"},"agent_actions":{"view_html":"https://pith.science/pith/K53GI27IDFN4JDN7JNDC3PAHCH","download_json":"https://pith.science/pith/K53GI27IDFN4JDN7JNDC3PAHCH.json","view_paper":"https://pith.science/paper/K53GI27I","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1511.06314&json=true","fetch_graph":"https://pith.science/api/pith-number/K53GI27IDFN4JDN7JNDC3PAHCH/graph.json","fetch_events":"https://pith.science/api/pith-number/K53GI27IDFN4JDN7JNDC3PAHCH/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/K53GI27IDFN4JDN7JNDC3PAHCH/action/timestamp_anchor","attest_storage":"https://pith.science/pith/K53GI27IDFN4JDN7JNDC3PAHCH/action/storage_attestation","attest_author":"https://pith.science/pith/K53GI27IDFN4JDN7JNDC3PAHCH/action/author_attestation","sign_citation":"https://pith.science/pith/K53GI27IDFN4JDN7JNDC3PAHCH/action/citation_signature","submit_replication":"https://pith.science/pith/K53GI27IDFN4JDN7JNDC3PAHCH/action/replication_record"}},"created_at":"2026-05-18T01:26:25.924298+00:00","updated_at":"2026-05-18T01:26:25.924298+00:00"}