{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:5U7KYT2BAUBPKWYEB4MNK3PZ2P","short_pith_number":"pith:5U7KYT2B","schema_version":"1.0","canonical_sha256":"ed3eac4f410502f55b040f18d56df9d3eed02ea8953b4671322a2c99cadd7eb5","source":{"kind":"arxiv","id":"1811.07755","version":2},"attestation_state":"computed","paper":{"title":"Building Efficient Deep Neural Networks with Unitary Group Convolutions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Christopher De Sa, Jordan Dotzel, Ritchie Zhao, Yuwei Hu, Zhiru Zhang","submitted_at":"2018-11-19T15:48:12Z","abstract_excerpt":"We propose unitary group convolutions (UGConvs), a building block for CNNs which compose a group convolution with unitary transforms in feature space to learn a richer set of representations than group convolution alone. UGConvs generalize two disparate ideas in CNN architecture, channel shuffling (i.e. ShuffleNet) and block-circulant networks (i.e. CirCNN), and provide unifying insights that lead to a deeper understanding of each technique. We experimentally demonstrate that dense unitary transforms can outperform channel shuffling in DNN accuracy. On the other hand, different dense transform"},"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":"1811.07755","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-11-19T15:48:12Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"1e29c133a9355f27cee7d108d687c40b798470ddba2d58855ecc7c2a7906d97f","abstract_canon_sha256":"f65db88b0feabc071af4555f337c7ada2c625487da0299b1a8b359fc3212a720"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:49:02.967213Z","signature_b64":"aOb4GSdy+TdjBXT0+W/XhJdhkK6ctrwY86HjeSXXg59aBelzZOkCDg4OgmUBontQkV8KoCZD4xHSLv7oMAqSBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"ed3eac4f410502f55b040f18d56df9d3eed02ea8953b4671322a2c99cadd7eb5","last_reissued_at":"2026-05-17T23:49:02.966592Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:49:02.966592Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Building Efficient Deep Neural Networks with Unitary Group Convolutions","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Christopher De Sa, Jordan Dotzel, Ritchie Zhao, Yuwei Hu, Zhiru Zhang","submitted_at":"2018-11-19T15:48:12Z","abstract_excerpt":"We propose unitary group convolutions (UGConvs), a building block for CNNs which compose a group convolution with unitary transforms in feature space to learn a richer set of representations than group convolution alone. UGConvs generalize two disparate ideas in CNN architecture, channel shuffling (i.e. ShuffleNet) and block-circulant networks (i.e. CirCNN), and provide unifying insights that lead to a deeper understanding of each technique. We experimentally demonstrate that dense unitary transforms can outperform channel shuffling in DNN accuracy. On the other hand, different dense transform"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.07755","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":""},"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":"1811.07755","created_at":"2026-05-17T23:49:02.966690+00:00"},{"alias_kind":"arxiv_version","alias_value":"1811.07755v2","created_at":"2026-05-17T23:49:02.966690+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.07755","created_at":"2026-05-17T23:49:02.966690+00:00"},{"alias_kind":"pith_short_12","alias_value":"5U7KYT2BAUBP","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_16","alias_value":"5U7KYT2BAUBPKWYE","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_8","alias_value":"5U7KYT2B","created_at":"2026-05-18T12:32:08.215937+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/5U7KYT2BAUBPKWYEB4MNK3PZ2P","json":"https://pith.science/pith/5U7KYT2BAUBPKWYEB4MNK3PZ2P.json","graph_json":"https://pith.science/api/pith-number/5U7KYT2BAUBPKWYEB4MNK3PZ2P/graph.json","events_json":"https://pith.science/api/pith-number/5U7KYT2BAUBPKWYEB4MNK3PZ2P/events.json","paper":"https://pith.science/paper/5U7KYT2B"},"agent_actions":{"view_html":"https://pith.science/pith/5U7KYT2BAUBPKWYEB4MNK3PZ2P","download_json":"https://pith.science/pith/5U7KYT2BAUBPKWYEB4MNK3PZ2P.json","view_paper":"https://pith.science/paper/5U7KYT2B","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1811.07755&json=true","fetch_graph":"https://pith.science/api/pith-number/5U7KYT2BAUBPKWYEB4MNK3PZ2P/graph.json","fetch_events":"https://pith.science/api/pith-number/5U7KYT2BAUBPKWYEB4MNK3PZ2P/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/5U7KYT2BAUBPKWYEB4MNK3PZ2P/action/timestamp_anchor","attest_storage":"https://pith.science/pith/5U7KYT2BAUBPKWYEB4MNK3PZ2P/action/storage_attestation","attest_author":"https://pith.science/pith/5U7KYT2BAUBPKWYEB4MNK3PZ2P/action/author_attestation","sign_citation":"https://pith.science/pith/5U7KYT2BAUBPKWYEB4MNK3PZ2P/action/citation_signature","submit_replication":"https://pith.science/pith/5U7KYT2BAUBPKWYEB4MNK3PZ2P/action/replication_record"}},"created_at":"2026-05-17T23:49:02.966690+00:00","updated_at":"2026-05-17T23:49:02.966690+00:00"}