{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2019:LGCXRJQCN3SMMZCHGLTY7G5TX4","short_pith_number":"pith:LGCXRJQC","canonical_record":{"source":{"id":"1910.10386","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-23T07:01:17Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"a873303fbb662c83ca5d9983fb4b17721a6ae1c90cccbb35fb38c27b06d3a8e4","abstract_canon_sha256":"50b357ed2d6b5bc981b78dd8b25132c8da3f188068fc28e9661fa8f6997b7c03"},"schema_version":"1.0"},"canonical_sha256":"598578a6026ee4c6644732e78f9bb3bf3fe88dfa140d9a78324d0e9f23cbad24","source":{"kind":"arxiv","id":"1910.10386","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1910.10386","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"arxiv_version","alias_value":"1910.10386v1","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.10386","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"pith_short_12","alias_value":"LGCXRJQCN3SM","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"pith_short_16","alias_value":"LGCXRJQCN3SMMZCH","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"pith_short_8","alias_value":"LGCXRJQC","created_at":"2026-07-05T00:14:24Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2019:LGCXRJQCN3SMMZCHGLTY7G5TX4","target":"record","payload":{"canonical_record":{"source":{"id":"1910.10386","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-23T07:01:17Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"a873303fbb662c83ca5d9983fb4b17721a6ae1c90cccbb35fb38c27b06d3a8e4","abstract_canon_sha256":"50b357ed2d6b5bc981b78dd8b25132c8da3f188068fc28e9661fa8f6997b7c03"},"schema_version":"1.0"},"canonical_sha256":"598578a6026ee4c6644732e78f9bb3bf3fe88dfa140d9a78324d0e9f23cbad24","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T00:14:24.518436Z","signature_b64":"p6XRe3SgqWCzJ87CAqhGn7h4QH7xSgBqDjH4EwRSa28ZWyydCUYGM+qkRgoQZg2u+2PzTD/+TnVazTEiyDB9AQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"598578a6026ee4c6644732e78f9bb3bf3fe88dfa140d9a78324d0e9f23cbad24","last_reissued_at":"2026-07-05T00:14:24.517877Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T00:14:24.517877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1910.10386","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":"2026-07-05T00:14:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"tZZNXGVNKQ72OVpKrthPnJN/ImRpXPI5VtYAQkkFebgOxdlpvPaU8s/dx9JmRMKsSP11GL3C8Ix+Itmh7OKzAg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T07:32:39.709889Z"},"content_sha256":"da9b4eb49532875dca43ef8a56c129dbc1be2456f1078041159e9379e70bae28","schema_version":"1.0","event_id":"sha256:da9b4eb49532875dca43ef8a56c129dbc1be2456f1078041159e9379e70bae28"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2019:LGCXRJQCN3SMMZCHGLTY7G5TX4","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Stabilising priors for robust Bayesian deep learning","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Arnu Pretorius, Felix McGregor, Johan du Preez, Steve Kroon","submitted_at":"2019-10-23T07:01:17Z","abstract_excerpt":"Bayesian neural networks (BNNs) have developed into useful tools for probabilistic modelling due to recent advances in variational inference enabling large scale BNNs. However, BNNs remain brittle and hard to train, especially: (1) when using deep architectures consisting of many hidden layers and (2) in situations with large weight variances. We use signal propagation theory to quantify these challenges and propose self-stabilising priors. This is achieved by a reformulation of the ELBO to allow the prior to influence network signal propagation. Then, we develop a stabilising prior, where the"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.10386","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/1910.10386/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":"2026-07-05T00:14:24Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"EB21VGq21SP0mbxmT9pYKSfdG3ay8MdxskTqDxxW2XDmJ2Ue26+pV4rRxMUpfWqTBv/GLcY13RHErxGMV58vCg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-07-06T07:32:39.710549Z"},"content_sha256":"f695f195a5bb5a703db0b7bc2528c65a9ddd40b10d610b4f5bfbe5e3d14ae6d2","schema_version":"1.0","event_id":"sha256:f695f195a5bb5a703db0b7bc2528c65a9ddd40b10d610b4f5bfbe5e3d14ae6d2"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/LGCXRJQCN3SMMZCHGLTY7G5TX4/bundle.json","state_url":"https://pith.science/pith/LGCXRJQCN3SMMZCHGLTY7G5TX4/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/LGCXRJQCN3SMMZCHGLTY7G5TX4/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":"2026-07-06T07:32:39Z","links":{"resolver":"https://pith.science/pith/LGCXRJQCN3SMMZCHGLTY7G5TX4","bundle":"https://pith.science/pith/LGCXRJQCN3SMMZCHGLTY7G5TX4/bundle.json","state":"https://pith.science/pith/LGCXRJQCN3SMMZCHGLTY7G5TX4/state.json","well_known_bundle":"https://pith.science/.well-known/pith/LGCXRJQCN3SMMZCHGLTY7G5TX4/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2019:LGCXRJQCN3SMMZCHGLTY7G5TX4","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":"50b357ed2d6b5bc981b78dd8b25132c8da3f188068fc28e9661fa8f6997b7c03","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-23T07:01:17Z","title_canon_sha256":"a873303fbb662c83ca5d9983fb4b17721a6ae1c90cccbb35fb38c27b06d3a8e4"},"schema_version":"1.0","source":{"id":"1910.10386","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1910.10386","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"arxiv_version","alias_value":"1910.10386v1","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1910.10386","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"pith_short_12","alias_value":"LGCXRJQCN3SM","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"pith_short_16","alias_value":"LGCXRJQCN3SMMZCH","created_at":"2026-07-05T00:14:24Z"},{"alias_kind":"pith_short_8","alias_value":"LGCXRJQC","created_at":"2026-07-05T00:14:24Z"}],"graph_snapshots":[{"event_id":"sha256:f695f195a5bb5a703db0b7bc2528c65a9ddd40b10d610b4f5bfbe5e3d14ae6d2","target":"graph","created_at":"2026-07-05T00:14:24Z","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/1910.10386/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Bayesian neural networks (BNNs) have developed into useful tools for probabilistic modelling due to recent advances in variational inference enabling large scale BNNs. However, BNNs remain brittle and hard to train, especially: (1) when using deep architectures consisting of many hidden layers and (2) in situations with large weight variances. We use signal propagation theory to quantify these challenges and propose self-stabilising priors. This is achieved by a reformulation of the ELBO to allow the prior to influence network signal propagation. Then, we develop a stabilising prior, where the","authors_text":"Arnu Pretorius, Felix McGregor, Johan du Preez, Steve Kroon","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-23T07:01:17Z","title":"Stabilising priors for robust Bayesian deep learning"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1910.10386","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:da9b4eb49532875dca43ef8a56c129dbc1be2456f1078041159e9379e70bae28","target":"record","created_at":"2026-07-05T00:14:24Z","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":"50b357ed2d6b5bc981b78dd8b25132c8da3f188068fc28e9661fa8f6997b7c03","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2019-10-23T07:01:17Z","title_canon_sha256":"a873303fbb662c83ca5d9983fb4b17721a6ae1c90cccbb35fb38c27b06d3a8e4"},"schema_version":"1.0","source":{"id":"1910.10386","kind":"arxiv","version":1}},"canonical_sha256":"598578a6026ee4c6644732e78f9bb3bf3fe88dfa140d9a78324d0e9f23cbad24","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"598578a6026ee4c6644732e78f9bb3bf3fe88dfa140d9a78324d0e9f23cbad24","first_computed_at":"2026-07-05T00:14:24.517877Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T00:14:24.517877Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"p6XRe3SgqWCzJ87CAqhGn7h4QH7xSgBqDjH4EwRSa28ZWyydCUYGM+qkRgoQZg2u+2PzTD/+TnVazTEiyDB9AQ==","signature_status":"signed_v1","signed_at":"2026-07-05T00:14:24.518436Z","signed_message":"canonical_sha256_bytes"},"source_id":"1910.10386","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:da9b4eb49532875dca43ef8a56c129dbc1be2456f1078041159e9379e70bae28","sha256:f695f195a5bb5a703db0b7bc2528c65a9ddd40b10d610b4f5bfbe5e3d14ae6d2"],"state_sha256":"4e9a708087fd60dd99242f6fbe8a530fd623682d11d107db26ca848cc27724fb"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"Z9hKh5UOYs3nDnAT4wPipdZ9vpjQhwXZt6VNMlDgbIQTQ92/PIzUu1b/389u1BRW1CYT/xh0gQYn6ELvlla5AA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-07-06T07:32:39.713508Z","bundle_sha256":"75967e564b2225f0f4c2d57a52aa20e3effb769b827fdb54b234157fe83ec635"}}