{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2017:Q6EIRBG4OKGUPOTZD5HNBOJ2BW","short_pith_number":"pith:Q6EIRBG4","canonical_record":{"source":{"id":"1704.08863","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-04-28T09:57:52Z","cross_cats_sorted":[],"title_canon_sha256":"1702f7f913d7f59640ee54bd8b87ddf2c6e95592df5028446239f450af2c2ef3","abstract_canon_sha256":"f92792d16203a01f7201286a644a3093292bf8a5b892f574727db6723344832f"},"schema_version":"1.0"},"canonical_sha256":"87888884dc728d47ba791f4ed0b93a0dadbf4e8266249250380b988174e814bc","source":{"kind":"arxiv","id":"1704.08863","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1704.08863","created_at":"2026-05-18T00:45:05Z"},{"alias_kind":"arxiv_version","alias_value":"1704.08863v2","created_at":"2026-05-18T00:45:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.08863","created_at":"2026-05-18T00:45:05Z"},{"alias_kind":"pith_short_12","alias_value":"Q6EIRBG4OKGU","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_16","alias_value":"Q6EIRBG4OKGUPOTZ","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_8","alias_value":"Q6EIRBG4","created_at":"2026-05-18T12:31:37Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2017:Q6EIRBG4OKGUPOTZD5HNBOJ2BW","target":"record","payload":{"canonical_record":{"source":{"id":"1704.08863","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-04-28T09:57:52Z","cross_cats_sorted":[],"title_canon_sha256":"1702f7f913d7f59640ee54bd8b87ddf2c6e95592df5028446239f450af2c2ef3","abstract_canon_sha256":"f92792d16203a01f7201286a644a3093292bf8a5b892f574727db6723344832f"},"schema_version":"1.0"},"canonical_sha256":"87888884dc728d47ba791f4ed0b93a0dadbf4e8266249250380b988174e814bc","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:45:05.980764Z","signature_b64":"2dftVI+YluiijvryBNxwDFMnYlltFTmmUgcgfxd2SCo58KkWe8qIDYlgOO8oT4TQs9uT7k5UPbR0POSWSMNMAA==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"87888884dc728d47ba791f4ed0b93a0dadbf4e8266249250380b988174e814bc","last_reissued_at":"2026-05-18T00:45:05.980350Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:45:05.980350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1704.08863","source_version":2,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:45:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"h+0++ciFArQgWptaaPDNWopBweWG0mI5RhRRQiyk3d+tUute+eNo+LPwx/TwC9Y5zexvRwVcE4eBA8XapmPaCw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T02:50:07.785913Z"},"content_sha256":"54faefa6c0f83765e23f43011c03c80b90076716bc5e377cd018e5603da86137","schema_version":"1.0","event_id":"sha256:54faefa6c0f83765e23f43011c03c80b90076716bc5e377cd018e5603da86137"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2017:Q6EIRBG4OKGUPOTZD5HNBOJ2BW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"On weight initialization in deep neural networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.LG","authors_text":"Siddharth Krishna Kumar","submitted_at":"2017-04-28T09:57:52Z","abstract_excerpt":"A proper initialization of the weights in a neural network is critical to its convergence. Current insights into weight initialization come primarily from linear activation functions. In this paper, I develop a theory for weight initializations with non-linear activations. First, I derive a general weight initialization strategy for any neural network using activation functions differentiable at 0. Next, I derive the weight initialization strategy for the Rectified Linear Unit (RELU), and provide theoretical insights into why the Xavier initialization is a poor choice with RELU activations. My"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.08863","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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:45:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"xQu6xaOSx0qeQeT1MtTcLfGqk5wdVsX5p6v69OxLlYnYHHKlN7QnDH3Ebtl/lk3vJWy2jKeRMAV5/J1i1AW3DQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-05T02:50:07.786611Z"},"content_sha256":"b11db3f9fbb0a0e1e2128918c35b62f34f40fa1a3ca725b3d2c82cfee92e7430","schema_version":"1.0","event_id":"sha256:b11db3f9fbb0a0e1e2128918c35b62f34f40fa1a3ca725b3d2c82cfee92e7430"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/Q6EIRBG4OKGUPOTZD5HNBOJ2BW/bundle.json","state_url":"https://pith.science/pith/Q6EIRBG4OKGUPOTZD5HNBOJ2BW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/Q6EIRBG4OKGUPOTZD5HNBOJ2BW/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":"2026-06-05T02:50:07Z","links":{"resolver":"https://pith.science/pith/Q6EIRBG4OKGUPOTZD5HNBOJ2BW","bundle":"https://pith.science/pith/Q6EIRBG4OKGUPOTZD5HNBOJ2BW/bundle.json","state":"https://pith.science/pith/Q6EIRBG4OKGUPOTZD5HNBOJ2BW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/Q6EIRBG4OKGUPOTZD5HNBOJ2BW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2017:Q6EIRBG4OKGUPOTZD5HNBOJ2BW","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":"f92792d16203a01f7201286a644a3093292bf8a5b892f574727db6723344832f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-04-28T09:57:52Z","title_canon_sha256":"1702f7f913d7f59640ee54bd8b87ddf2c6e95592df5028446239f450af2c2ef3"},"schema_version":"1.0","source":{"id":"1704.08863","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1704.08863","created_at":"2026-05-18T00:45:05Z"},{"alias_kind":"arxiv_version","alias_value":"1704.08863v2","created_at":"2026-05-18T00:45:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1704.08863","created_at":"2026-05-18T00:45:05Z"},{"alias_kind":"pith_short_12","alias_value":"Q6EIRBG4OKGU","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_16","alias_value":"Q6EIRBG4OKGUPOTZ","created_at":"2026-05-18T12:31:37Z"},{"alias_kind":"pith_short_8","alias_value":"Q6EIRBG4","created_at":"2026-05-18T12:31:37Z"}],"graph_snapshots":[{"event_id":"sha256:b11db3f9fbb0a0e1e2128918c35b62f34f40fa1a3ca725b3d2c82cfee92e7430","target":"graph","created_at":"2026-05-18T00:45:05Z","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"},"paper":{"abstract_excerpt":"A proper initialization of the weights in a neural network is critical to its convergence. Current insights into weight initialization come primarily from linear activation functions. In this paper, I develop a theory for weight initializations with non-linear activations. First, I derive a general weight initialization strategy for any neural network using activation functions differentiable at 0. Next, I derive the weight initialization strategy for the Rectified Linear Unit (RELU), and provide theoretical insights into why the Xavier initialization is a poor choice with RELU activations. My","authors_text":"Siddharth Krishna Kumar","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-04-28T09:57:52Z","title":"On weight initialization in deep neural networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1704.08863","kind":"arxiv","version":2},"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":"sha256:54faefa6c0f83765e23f43011c03c80b90076716bc5e377cd018e5603da86137","target":"record","created_at":"2026-05-18T00:45:05Z","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":"f92792d16203a01f7201286a644a3093292bf8a5b892f574727db6723344832f","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2017-04-28T09:57:52Z","title_canon_sha256":"1702f7f913d7f59640ee54bd8b87ddf2c6e95592df5028446239f450af2c2ef3"},"schema_version":"1.0","source":{"id":"1704.08863","kind":"arxiv","version":2}},"canonical_sha256":"87888884dc728d47ba791f4ed0b93a0dadbf4e8266249250380b988174e814bc","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"87888884dc728d47ba791f4ed0b93a0dadbf4e8266249250380b988174e814bc","first_computed_at":"2026-05-18T00:45:05.980350Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:45:05.980350Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"2dftVI+YluiijvryBNxwDFMnYlltFTmmUgcgfxd2SCo58KkWe8qIDYlgOO8oT4TQs9uT7k5UPbR0POSWSMNMAA==","signature_status":"signed_v1","signed_at":"2026-05-18T00:45:05.980764Z","signed_message":"canonical_sha256_bytes"},"source_id":"1704.08863","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:54faefa6c0f83765e23f43011c03c80b90076716bc5e377cd018e5603da86137","sha256:b11db3f9fbb0a0e1e2128918c35b62f34f40fa1a3ca725b3d2c82cfee92e7430"],"state_sha256":"ff107130af464c97a44ba8617fc80051a43f75a8f79a77b6f53358b5878817a0"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"y3SUTOSp42d7J8oUUycGeX1hscflwXtI1CKaWHKX5xN26lpDsFDbSjW4zLS11dU6XJwsSPF7NVHFKO6KUcp9Aw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-05T02:50:07.790409Z","bundle_sha256":"991e1e9a361d1db27ba110381baac78b827984fb841c16bd3e269e4c90356ff9"}}