{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2018:6DFKPMMQUA7GP4KK5K7REXH4IU","short_pith_number":"pith:6DFKPMMQ","schema_version":"1.0","canonical_sha256":"f0caa7b190a03e67f14aeabf125cfc451e05e4e5bb77c79700bd33ae49ae647e","source":{"kind":"arxiv","id":"1811.10779","version":1},"attestation_state":"computed","paper":{"title":"Tackling Early Sparse Gradients in Softmax Activation Using Leaky Squared Euclidean Distance","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Rujie Liu, Wei Shen","submitted_at":"2018-11-27T02:37:03Z","abstract_excerpt":"Softmax activation is commonly used to output the probability distribution over categories based on certain distance metric. In scenarios like one-shot learning, the distance metric is often chosen to be squared Euclidean distance between the query sample and the category prototype. This practice works well in most time. However, we find that choosing squared Euclidean distance may cause distance explosion leading gradients to be extremely sparse in the early stage of back propagation. We term this phenomena as the early sparse gradients problem. Though it doesn't deteriorate the convergence o"},"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":"1811.10779","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-11-27T02:37:03Z","cross_cats_sorted":[],"title_canon_sha256":"21334485c10c3275276ad653b5eed50889a410a6d686027623f818e809d377b1","abstract_canon_sha256":"46c13d09618021ce34e3f18170197e5ef0393e8b6ca3fd079f37e6cad54077bd"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:59:47.553788Z","signature_b64":"JYn48/yy1fdm5cZQD2AHx34ILQYrHKKuwmYXs7xsceveNsYzJ7Mfgi95t6EF/HAyzq2634an+BAy2pZWBMLgBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"f0caa7b190a03e67f14aeabf125cfc451e05e4e5bb77c79700bd33ae49ae647e","last_reissued_at":"2026-05-17T23:59:47.553376Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:59:47.553376Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Tackling Early Sparse Gradients in Softmax Activation Using Leaky Squared Euclidean Distance","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Rujie Liu, Wei Shen","submitted_at":"2018-11-27T02:37:03Z","abstract_excerpt":"Softmax activation is commonly used to output the probability distribution over categories based on certain distance metric. In scenarios like one-shot learning, the distance metric is often chosen to be squared Euclidean distance between the query sample and the category prototype. This practice works well in most time. However, we find that choosing squared Euclidean distance may cause distance explosion leading gradients to be extremely sparse in the early stage of back propagation. We term this phenomena as the early sparse gradients problem. Though it doesn't deteriorate the convergence o"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.10779","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":"1811.10779","created_at":"2026-05-17T23:59:47.553464+00:00"},{"alias_kind":"arxiv_version","alias_value":"1811.10779v1","created_at":"2026-05-17T23:59:47.553464+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.10779","created_at":"2026-05-17T23:59:47.553464+00:00"},{"alias_kind":"pith_short_12","alias_value":"6DFKPMMQUA7G","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_16","alias_value":"6DFKPMMQUA7GP4KK","created_at":"2026-05-18T12:32:08.215937+00:00"},{"alias_kind":"pith_short_8","alias_value":"6DFKPMMQ","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/6DFKPMMQUA7GP4KK5K7REXH4IU","json":"https://pith.science/pith/6DFKPMMQUA7GP4KK5K7REXH4IU.json","graph_json":"https://pith.science/api/pith-number/6DFKPMMQUA7GP4KK5K7REXH4IU/graph.json","events_json":"https://pith.science/api/pith-number/6DFKPMMQUA7GP4KK5K7REXH4IU/events.json","paper":"https://pith.science/paper/6DFKPMMQ"},"agent_actions":{"view_html":"https://pith.science/pith/6DFKPMMQUA7GP4KK5K7REXH4IU","download_json":"https://pith.science/pith/6DFKPMMQUA7GP4KK5K7REXH4IU.json","view_paper":"https://pith.science/paper/6DFKPMMQ","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1811.10779&json=true","fetch_graph":"https://pith.science/api/pith-number/6DFKPMMQUA7GP4KK5K7REXH4IU/graph.json","fetch_events":"https://pith.science/api/pith-number/6DFKPMMQUA7GP4KK5K7REXH4IU/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/6DFKPMMQUA7GP4KK5K7REXH4IU/action/timestamp_anchor","attest_storage":"https://pith.science/pith/6DFKPMMQUA7GP4KK5K7REXH4IU/action/storage_attestation","attest_author":"https://pith.science/pith/6DFKPMMQUA7GP4KK5K7REXH4IU/action/author_attestation","sign_citation":"https://pith.science/pith/6DFKPMMQUA7GP4KK5K7REXH4IU/action/citation_signature","submit_replication":"https://pith.science/pith/6DFKPMMQUA7GP4KK5K7REXH4IU/action/replication_record"}},"created_at":"2026-05-17T23:59:47.553464+00:00","updated_at":"2026-05-17T23:59:47.553464+00:00"}