{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:24YGYHB3PSJ4PIZHOLY42T3IZM","short_pith_number":"pith:24YGYHB3","canonical_record":{"source":{"id":"2102.09393","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2021-02-18T14:37:25Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"fc3fb27ac74c142dc37e3b41c2174f75325e97f3f0ba64c76065b3e89cf0f049","abstract_canon_sha256":"80f97b2ef6c7e2c2b0d85268e6cc76565608d9d015c73afbf9b7d8e9b938d88f"},"schema_version":"1.0"},"canonical_sha256":"d7306c1c3b7c93c7a32772f1cd4f68cb24268f3fdc41ce8bc372e013ca6244e3","source":{"kind":"arxiv","id":"2102.09393","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2102.09393","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"arxiv_version","alias_value":"2102.09393v1","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.09393","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"pith_short_12","alias_value":"24YGYHB3PSJ4","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"pith_short_16","alias_value":"24YGYHB3PSJ4PIZH","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"pith_short_8","alias_value":"24YGYHB3","created_at":"2026-07-05T02:16:22Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:24YGYHB3PSJ4PIZHOLY42T3IZM","target":"record","payload":{"canonical_record":{"source":{"id":"2102.09393","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2021-02-18T14:37:25Z","cross_cats_sorted":["cs.LG","stat.ML"],"title_canon_sha256":"fc3fb27ac74c142dc37e3b41c2174f75325e97f3f0ba64c76065b3e89cf0f049","abstract_canon_sha256":"80f97b2ef6c7e2c2b0d85268e6cc76565608d9d015c73afbf9b7d8e9b938d88f"},"schema_version":"1.0"},"canonical_sha256":"d7306c1c3b7c93c7a32772f1cd4f68cb24268f3fdc41ce8bc372e013ca6244e3","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T02:16:22.598414Z","signature_b64":"KQHZynNPMnSl+HsMBzBSpIXf9InSfq+XsHkZXQi4htb0Cp2BNNKHnrzrXZCwBTQQ8oeMdf2A2aW5pYj4CloMAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d7306c1c3b7c93c7a32772f1cd4f68cb24268f3fdc41ce8bc372e013ca6244e3","last_reissued_at":"2026-07-05T02:16:22.597976Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T02:16:22.597976Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2102.09393","source_version":1,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:16:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"k6OysuC0IPFNUy8JtqYbOHL6aElqTe9/89GGnGDG3cWVbo9/wcuWKKvu9aA2zIALBLKIgi12kd9444dLM95SCQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T00:06:02.249429Z"},"content_sha256":"85143ee9305d563ac69c7fc372eadd90890389f580df27427ebccbb0d7951d57","schema_version":"1.0","event_id":"sha256:85143ee9305d563ac69c7fc372eadd90890389f580df27427ebccbb0d7951d57"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:24YGYHB3PSJ4PIZHOLY42T3IZM","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On the Convergence of Step Decay Step-Size for Stochastic Optimization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG","stat.ML"],"primary_cat":"math.OC","authors_text":"Mikael Johansson, Sindri Magn\\'usson, Xiaoyu Wang","submitted_at":"2021-02-18T14:37:25Z","abstract_excerpt":"The convergence of stochastic gradient descent is highly dependent on the step-size, especially on non-convex problems such as neural network training. Step decay step-size schedules (constant and then cut) are widely used in practice because of their excellent convergence and generalization qualities, but their theoretical properties are not yet well understood. We provide the convergence results for step decay in the non-convex regime, ensuring that the gradient norm vanishes at an $\\mathcal{O}(\\ln T/\\sqrt{T})$ rate. We also provide the convergence guarantees for general (possibly non-smooth"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.09393","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2102.09393/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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T02:16:22Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"iwRgKfiSS2pw/sygZNRtD4PE427zB0Nv1vr+rSIcSq0VvjrJTJybaQACdviae4K/2+4KMLemPX/KcjgxKq53DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-19T00:06:02.249941Z"},"content_sha256":"1bb38bb22f011f4e4d134878c18a038d17d580466855533b32cd52dc3303abcd","schema_version":"1.0","event_id":"sha256:1bb38bb22f011f4e4d134878c18a038d17d580466855533b32cd52dc3303abcd"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/24YGYHB3PSJ4PIZHOLY42T3IZM/bundle.json","state_url":"https://pith.science/pith/24YGYHB3PSJ4PIZHOLY42T3IZM/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/24YGYHB3PSJ4PIZHOLY42T3IZM/bundle.json","status":"primary"}],"public_keys":[{"key_id":"pith-v1-2026-05","algorithm":"ed25519","format":"raw","public_key_b64":"stVStoiQhXFxp4s2pdzPNoqVNBMojDU/fJ2db5S3CbM=","public_key_hex":"b2d552b68890857171a78b36a5dccf368a953413288c353f7c9d9d6f94b709b3","fingerprint_sha256_b32_first128bits":"RVFV5Z2OI2J3ZUO7ERDEBCYNKS","fingerprint_sha256_hex":"8d4b5ee74e4693bcd1df2446408b0d54","rotates_at":null,"url":"https://pith.science/pith-signing-key.json","notes":"Pith uses this Ed25519 key to sign canonical record SHA-256 digests. Verify with: ed25519_verify(public_key, message=canonical_sha256_bytes, signature=base64decode(signature_b64))."}],"merge_version":"pith-open-graph-merge-v1","built_at":"2026-08-19T00:06:02Z","links":{"resolver":"https://pith.science/pith/24YGYHB3PSJ4PIZHOLY42T3IZM","bundle":"https://pith.science/pith/24YGYHB3PSJ4PIZHOLY42T3IZM/bundle.json","state":"https://pith.science/pith/24YGYHB3PSJ4PIZHOLY42T3IZM/state.json","well_known_bundle":"https://pith.science/.well-known/pith/24YGYHB3PSJ4PIZHOLY42T3IZM/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:24YGYHB3PSJ4PIZHOLY42T3IZM","merge_version":"pith-open-graph-merge-v1","event_count":2,"valid_event_count":2,"invalid_event_count":0,"equivocation_count":0,"current":{"canonical_record":{"metadata":{"abstract_canon_sha256":"80f97b2ef6c7e2c2b0d85268e6cc76565608d9d015c73afbf9b7d8e9b938d88f","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2021-02-18T14:37:25Z","title_canon_sha256":"fc3fb27ac74c142dc37e3b41c2174f75325e97f3f0ba64c76065b3e89cf0f049"},"schema_version":"1.0","source":{"id":"2102.09393","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2102.09393","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"arxiv_version","alias_value":"2102.09393v1","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2102.09393","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"pith_short_12","alias_value":"24YGYHB3PSJ4","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"pith_short_16","alias_value":"24YGYHB3PSJ4PIZH","created_at":"2026-07-05T02:16:22Z"},{"alias_kind":"pith_short_8","alias_value":"24YGYHB3","created_at":"2026-07-05T02:16:22Z"}],"graph_snapshots":[{"event_id":"sha256:1bb38bb22f011f4e4d134878c18a038d17d580466855533b32cd52dc3303abcd","target":"graph","created_at":"2026-07-05T02:16:22Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"graph_snapshot":{"author_claims":{"count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57","strong_count":0},"builder_version":"pith-number-builder-2026-05-17-v1","claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"formal_canon":{"evidence_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"integrity":{"available":true,"clean":true,"detectors_run":[],"endpoint":"/pith/2102.09393/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"The convergence of stochastic gradient descent is highly dependent on the step-size, especially on non-convex problems such as neural network training. Step decay step-size schedules (constant and then cut) are widely used in practice because of their excellent convergence and generalization qualities, but their theoretical properties are not yet well understood. We provide the convergence results for step decay in the non-convex regime, ensuring that the gradient norm vanishes at an $\\mathcal{O}(\\ln T/\\sqrt{T})$ rate. We also provide the convergence guarantees for general (possibly non-smooth","authors_text":"Mikael Johansson, Sindri Magn\\'usson, Xiaoyu Wang","cross_cats":["cs.LG","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2021-02-18T14:37:25Z","title":"On the Convergence of Step Decay Step-Size for Stochastic Optimization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2102.09393","kind":"arxiv","version":1},"verdict":{"created_at":null,"id":null,"model_set":{},"one_line_summary":"","pipeline_version":null,"pith_extraction_headline":"","strongest_claim":"","weakest_assumption":""}},"verdict_id":null}}],"author_attestations":[],"timestamp_anchors":[],"storage_attestations":[],"citation_signatures":[],"replication_records":[],"corrections":[],"mirror_hints":[],"record_created":{"event_id":"sha256:85143ee9305d563ac69c7fc372eadd90890389f580df27427ebccbb0d7951d57","target":"record","created_at":"2026-07-05T02:16:22Z","signer":{"key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signer_id":"pith.science","signer_type":"pith_registry"},"payload":{"attestation_state":"computed","canonical_record":{"metadata":{"abstract_canon_sha256":"80f97b2ef6c7e2c2b0d85268e6cc76565608d9d015c73afbf9b7d8e9b938d88f","cross_cats_sorted":["cs.LG","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"math.OC","submitted_at":"2021-02-18T14:37:25Z","title_canon_sha256":"fc3fb27ac74c142dc37e3b41c2174f75325e97f3f0ba64c76065b3e89cf0f049"},"schema_version":"1.0","source":{"id":"2102.09393","kind":"arxiv","version":1}},"canonical_sha256":"d7306c1c3b7c93c7a32772f1cd4f68cb24268f3fdc41ce8bc372e013ca6244e3","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d7306c1c3b7c93c7a32772f1cd4f68cb24268f3fdc41ce8bc372e013ca6244e3","first_computed_at":"2026-07-05T02:16:22.597976Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T02:16:22.597976Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"KQHZynNPMnSl+HsMBzBSpIXf9InSfq+XsHkZXQi4htb0Cp2BNNKHnrzrXZCwBTQQ8oeMdf2A2aW5pYj4CloMAw==","signature_status":"signed_v1","signed_at":"2026-07-05T02:16:22.598414Z","signed_message":"canonical_sha256_bytes"},"source_id":"2102.09393","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:85143ee9305d563ac69c7fc372eadd90890389f580df27427ebccbb0d7951d57","sha256:1bb38bb22f011f4e4d134878c18a038d17d580466855533b32cd52dc3303abcd"],"state_sha256":"e2381fd196a4649b58f2ee4751caed0223080f21630a0da97a8659278c7d191f"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Bz8di9DU7IFp2oA+HYEEXTlfuvNtf9n5DRR2q8UZbhnp91rADTEEzNoJ7oq6ugdJ72oZNtImdFia4GwyCKkfCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-19T00:06:02.255645Z","bundle_sha256":"23313fdffaf59c02afb585b7f901fe432d06feb96b1fae86e7ce191a1a332558"}}