{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2017:2V76GQQFJAVGAKZQRMSJ4B5TGP","short_pith_number":"pith:2V76GQQF","schema_version":"1.0","canonical_sha256":"d57fe34205482a602b308b249e07b333f837b7db3a92e63ad895701243c2f6c0","source":{"kind":"arxiv","id":"1705.00607","version":2},"attestation_state":"computed","paper":{"title":"Determinantal Point Processes for Mini-Batch Diversification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Cheng Zhang, Hedvig Kjellstrom, Stephan Mandt","submitted_at":"2017-05-01T17:53:51Z","abstract_excerpt":"We study a mini-batch diversification scheme for stochastic gradient descent (SGD). While classical SGD relies on uniformly sampling data points to form a mini-batch, we propose a non-uniform sampling scheme based on the Determinantal Point Process (DPP). The DPP relies on a similarity measure between data points and gives low probabilities to mini-batches which contain redundant data, and higher probabilities to mini-batches with more diverse data. This simultaneously balances the data and leads to stochastic gradients with lower variance. We term this approach Diversified Mini-Batch SGD (DM-"},"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":"1705.00607","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-05-01T17:53:51Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"5763481a6a9cfcc6fd1cc1c187246e36690a0a7f5a096bd8568c94210509f7b4","abstract_canon_sha256":"40af75207435e39ba057231164cb87accedbe0811f4dc7b45a00bc6d074d76e7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:35:41.374476Z","signature_b64":"vq/lVWnE69BdUhxC4MjRWs1HPNu8LevKt6du4ThIdF+04BJT9lP94ICLiWhSv6YYzf4P781xmIJ1V0yE5sQjBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d57fe34205482a602b308b249e07b333f837b7db3a92e63ad895701243c2f6c0","last_reissued_at":"2026-05-18T00:35:41.373931Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:35:41.373931Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Determinantal Point Processes for Mini-Batch Diversification","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Cheng Zhang, Hedvig Kjellstrom, Stephan Mandt","submitted_at":"2017-05-01T17:53:51Z","abstract_excerpt":"We study a mini-batch diversification scheme for stochastic gradient descent (SGD). While classical SGD relies on uniformly sampling data points to form a mini-batch, we propose a non-uniform sampling scheme based on the Determinantal Point Process (DPP). The DPP relies on a similarity measure between data points and gives low probabilities to mini-batches which contain redundant data, and higher probabilities to mini-batches with more diverse data. This simultaneously balances the data and leads to stochastic gradients with lower variance. We term this approach Diversified Mini-Batch SGD (DM-"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1705.00607","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":"1705.00607","created_at":"2026-05-18T00:35:41.374015+00:00"},{"alias_kind":"arxiv_version","alias_value":"1705.00607v2","created_at":"2026-05-18T00:35:41.374015+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1705.00607","created_at":"2026-05-18T00:35:41.374015+00:00"},{"alias_kind":"pith_short_12","alias_value":"2V76GQQFJAVG","created_at":"2026-05-18T12:30:55.937587+00:00"},{"alias_kind":"pith_short_16","alias_value":"2V76GQQFJAVGAKZQ","created_at":"2026-05-18T12:30:55.937587+00:00"},{"alias_kind":"pith_short_8","alias_value":"2V76GQQF","created_at":"2026-05-18T12:30:55.937587+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":3,"internal_anchor_count":1,"sample":[{"citing_arxiv_id":"1906.08771","citing_title":"Submodular Batch Selection for Training Deep Neural Networks","ref_index":29,"is_internal_anchor":true},{"citing_arxiv_id":"2605.08843","citing_title":"M$^3$: Reframing Training Measures for Discretized Physical Simulations","ref_index":20,"is_internal_anchor":false},{"citing_arxiv_id":"2604.06350","citing_title":"Convergence of Riemannian Stochastic Gradient Descents: Varying Batch Sizes And Nonstandard Batch Forming","ref_index":28,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/2V76GQQFJAVGAKZQRMSJ4B5TGP","json":"https://pith.science/pith/2V76GQQFJAVGAKZQRMSJ4B5TGP.json","graph_json":"https://pith.science/api/pith-number/2V76GQQFJAVGAKZQRMSJ4B5TGP/graph.json","events_json":"https://pith.science/api/pith-number/2V76GQQFJAVGAKZQRMSJ4B5TGP/events.json","paper":"https://pith.science/paper/2V76GQQF"},"agent_actions":{"view_html":"https://pith.science/pith/2V76GQQFJAVGAKZQRMSJ4B5TGP","download_json":"https://pith.science/pith/2V76GQQFJAVGAKZQRMSJ4B5TGP.json","view_paper":"https://pith.science/paper/2V76GQQF","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=1705.00607&json=true","fetch_graph":"https://pith.science/api/pith-number/2V76GQQFJAVGAKZQRMSJ4B5TGP/graph.json","fetch_events":"https://pith.science/api/pith-number/2V76GQQFJAVGAKZQRMSJ4B5TGP/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/2V76GQQFJAVGAKZQRMSJ4B5TGP/action/timestamp_anchor","attest_storage":"https://pith.science/pith/2V76GQQFJAVGAKZQRMSJ4B5TGP/action/storage_attestation","attest_author":"https://pith.science/pith/2V76GQQFJAVGAKZQRMSJ4B5TGP/action/author_attestation","sign_citation":"https://pith.science/pith/2V76GQQFJAVGAKZQRMSJ4B5TGP/action/citation_signature","submit_replication":"https://pith.science/pith/2V76GQQFJAVGAKZQRMSJ4B5TGP/action/replication_record"}},"created_at":"2026-05-18T00:35:41.374015+00:00","updated_at":"2026-05-18T00:35:41.374015+00:00"}