{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2021:HTOYRGLZWGOIWIT5F62CYUY2AY","short_pith_number":"pith:HTOYRGLZ","schema_version":"1.0","canonical_sha256":"3cdd889979b19c8b227d2fb42c531a0636da7a0f90b76e093b9cbfd906bc3c95","source":{"kind":"arxiv","id":"2103.01148","version":3},"attestation_state":"computed","paper":{"title":"E$^2$CM: Early Exit via Class Means for Efficient Supervised and Unsupervised Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alperen G\\\"ormez, Erdem Koyuncu, Venkat R. Dasari","submitted_at":"2021-03-01T17:31:55Z","abstract_excerpt":"State-of-the-art neural networks with early exit mechanisms often need considerable amount of training and fine tuning to achieve good performance with low computational cost. We propose a novel early exit technique, Early Exit Class Means (E$^2$CM), based on class means of samples. Unlike most existing schemes, E$^2$CM does not require gradient-based training of internal classifiers and it does not modify the base network by any means. This makes it particularly useful for neural network training in low-power devices, as in wireless edge networks. We evaluate the performance and overheads of "},"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":"2103.01148","kind":"arxiv","version":3},"metadata":{"license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","primary_cat":"cs.LG","submitted_at":"2021-03-01T17:31:55Z","cross_cats_sorted":["cs.NE","stat.ML"],"title_canon_sha256":"0689b2e92dc925660963da2e77321b73a465b41071cc9793ca1ca9e0b7841d0d","abstract_canon_sha256":"ae9d9936f8590a93702ac47c9ccc68ab4bdea4021cc4b231af5f282e03846352"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:10:52.123528Z","signature_b64":"ZkHXvfQp1EfHCMkcswS3TvX8sFtoY2xzvzfkUNRLbHNNx3eYl/HxFh1UFYtbcMJVJyI0X4S0PiqvHHpnzSuzCw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"3cdd889979b19c8b227d2fb42c531a0636da7a0f90b76e093b9cbfd906bc3c95","last_reissued_at":"2026-07-05T05:10:52.123030Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:10:52.123030Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"E$^2$CM: Early Exit via Class Means for Efficient Supervised and Unsupervised Learning","license":"http://creativecommons.org/licenses/by-nc-sa/4.0/","headline":"","cross_cats":["cs.NE","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alperen G\\\"ormez, Erdem Koyuncu, Venkat R. Dasari","submitted_at":"2021-03-01T17:31:55Z","abstract_excerpt":"State-of-the-art neural networks with early exit mechanisms often need considerable amount of training and fine tuning to achieve good performance with low computational cost. We propose a novel early exit technique, Early Exit Class Means (E$^2$CM), based on class means of samples. Unlike most existing schemes, E$^2$CM does not require gradient-based training of internal classifiers and it does not modify the base network by any means. This makes it particularly useful for neural network training in low-power devices, as in wireless edge networks. We evaluate the performance and overheads of "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2103.01148","kind":"arxiv","version":3},"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/2103.01148/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":"2103.01148","created_at":"2026-07-05T05:10:52.123094+00:00"},{"alias_kind":"arxiv_version","alias_value":"2103.01148v3","created_at":"2026-07-05T05:10:52.123094+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2103.01148","created_at":"2026-07-05T05:10:52.123094+00:00"},{"alias_kind":"pith_short_12","alias_value":"HTOYRGLZWGOI","created_at":"2026-07-05T05:10:52.123094+00:00"},{"alias_kind":"pith_short_16","alias_value":"HTOYRGLZWGOIWIT5","created_at":"2026-07-05T05:10:52.123094+00:00"},{"alias_kind":"pith_short_8","alias_value":"HTOYRGLZ","created_at":"2026-07-05T05:10:52.123094+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":1,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"2412.19325","citing_title":"Performance Control in Early Exiting to Deploy Large Models at the Same Cost of Smaller Ones","ref_index":16,"is_internal_anchor":true}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/HTOYRGLZWGOIWIT5F62CYUY2AY","json":"https://pith.science/pith/HTOYRGLZWGOIWIT5F62CYUY2AY.json","graph_json":"https://pith.science/api/pith-number/HTOYRGLZWGOIWIT5F62CYUY2AY/graph.json","events_json":"https://pith.science/api/pith-number/HTOYRGLZWGOIWIT5F62CYUY2AY/events.json","paper":"https://pith.science/paper/HTOYRGLZ"},"agent_actions":{"view_html":"https://pith.science/pith/HTOYRGLZWGOIWIT5F62CYUY2AY","download_json":"https://pith.science/pith/HTOYRGLZWGOIWIT5F62CYUY2AY.json","view_paper":"https://pith.science/paper/HTOYRGLZ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2103.01148&json=true","fetch_graph":"https://pith.science/api/pith-number/HTOYRGLZWGOIWIT5F62CYUY2AY/graph.json","fetch_events":"https://pith.science/api/pith-number/HTOYRGLZWGOIWIT5F62CYUY2AY/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/HTOYRGLZWGOIWIT5F62CYUY2AY/action/timestamp_anchor","attest_storage":"https://pith.science/pith/HTOYRGLZWGOIWIT5F62CYUY2AY/action/storage_attestation","attest_author":"https://pith.science/pith/HTOYRGLZWGOIWIT5F62CYUY2AY/action/author_attestation","sign_citation":"https://pith.science/pith/HTOYRGLZWGOIWIT5F62CYUY2AY/action/citation_signature","submit_replication":"https://pith.science/pith/HTOYRGLZWGOIWIT5F62CYUY2AY/action/replication_record"}},"created_at":"2026-07-05T05:10:52.123094+00:00","updated_at":"2026-07-05T05:10:52.123094+00:00"}