{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:RFLE26KGE3X2VPU6ZC2YCUSAYW","short_pith_number":"pith:RFLE26KG","canonical_record":{"source":{"id":"1806.10317","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-27T06:24:53Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"b884c31331805090c034cbcdb47c36a4919a9804050dbc77c390daef8d08b61b","abstract_canon_sha256":"8662b9b659273fab50a8d828a2af48bcce418a7600e221c21cc7d2f57fe6d887"},"schema_version":"1.0"},"canonical_sha256":"89564d794626efaabe9ec8b5815240c58c6bf01eeb9807d389fd30984d8a1e2a","source":{"kind":"arxiv","id":"1806.10317","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.10317","created_at":"2026-05-18T00:12:12Z"},{"alias_kind":"arxiv_version","alias_value":"1806.10317v1","created_at":"2026-05-18T00:12:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.10317","created_at":"2026-05-18T00:12:12Z"},{"alias_kind":"pith_short_12","alias_value":"RFLE26KGE3X2","created_at":"2026-05-18T12:32:50Z"},{"alias_kind":"pith_short_16","alias_value":"RFLE26KGE3X2VPU6","created_at":"2026-05-18T12:32:50Z"},{"alias_kind":"pith_short_8","alias_value":"RFLE26KG","created_at":"2026-05-18T12:32:50Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:RFLE26KGE3X2VPU6ZC2YCUSAYW","target":"record","payload":{"canonical_record":{"source":{"id":"1806.10317","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-27T06:24:53Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"b884c31331805090c034cbcdb47c36a4919a9804050dbc77c390daef8d08b61b","abstract_canon_sha256":"8662b9b659273fab50a8d828a2af48bcce418a7600e221c21cc7d2f57fe6d887"},"schema_version":"1.0"},"canonical_sha256":"89564d794626efaabe9ec8b5815240c58c6bf01eeb9807d389fd30984d8a1e2a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:12:12.835354Z","signature_b64":"kQQ0SASz7BEfml9H50N8vysUNwLYljI155turMriWofzvF6wmNGo8LHQ8LQE9PTfejIIRd2p0vg0IKZ9h0atBw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"89564d794626efaabe9ec8b5815240c58c6bf01eeb9807d389fd30984d8a1e2a","last_reissued_at":"2026-05-18T00:12:12.834623Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:12:12.834623Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1806.10317","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":"2026-05-18T00:12:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"F94y6dXx/zZ7E/HSQbDNkNNkgUFHWcfFPDTUkJsqucpHAqCinImJNWxckRjF1rkfLlMiFYkJE40RphqEGkwOAw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T21:23:05.029269Z"},"content_sha256":"9ec5f2e35dbb819518b3bb3baeaf3558ded6ab4a35c722018df6e7c737d060be","schema_version":"1.0","event_id":"sha256:9ec5f2e35dbb819518b3bb3baeaf3558ded6ab4a35c722018df6e7c737d060be"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:RFLE26KGE3X2VPU6ZC2YCUSAYW","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Adversarial Distillation of Bayesian Neural Network Posteriors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"James Lucas, Kuan-Chieh Wang, Li Gu, Paul Vicol, Richard Zemel, Roger Grosse","submitted_at":"2018-06-27T06:24:53Z","abstract_excerpt":"Bayesian neural networks (BNNs) allow us to reason about uncertainty in a principled way. Stochastic Gradient Langevin Dynamics (SGLD) enables efficient BNN learning by drawing samples from the BNN posterior using mini-batches. However, SGLD and its extensions require storage of many copies of the model parameters, a potentially prohibitive cost, especially for large neural networks. We propose a framework, Adversarial Posterior Distillation, to distill the SGLD samples using a Generative Adversarial Network (GAN). At test-time, samples are generated by the GAN. We show that this distillation "},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.10317","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"},"verdict_id":null},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-05-18T00:12:12Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"oxhYDkv+bzgrCCSoa3LrpIXzJ/pyGyH9iwss1GOGoPJDp3CVWNdDKbDJceNEXbmppBmlpx68V/lNVuxvDQ3+BQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-31T21:23:05.029627Z"},"content_sha256":"14bd3d584742b389ce510a63d2403869f9035fdd6d76e96046c697f7e43cf9d8","schema_version":"1.0","event_id":"sha256:14bd3d584742b389ce510a63d2403869f9035fdd6d76e96046c697f7e43cf9d8"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/RFLE26KGE3X2VPU6ZC2YCUSAYW/bundle.json","state_url":"https://pith.science/pith/RFLE26KGE3X2VPU6ZC2YCUSAYW/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/RFLE26KGE3X2VPU6ZC2YCUSAYW/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":"2026-05-31T21:23:05Z","links":{"resolver":"https://pith.science/pith/RFLE26KGE3X2VPU6ZC2YCUSAYW","bundle":"https://pith.science/pith/RFLE26KGE3X2VPU6ZC2YCUSAYW/bundle.json","state":"https://pith.science/pith/RFLE26KGE3X2VPU6ZC2YCUSAYW/state.json","well_known_bundle":"https://pith.science/.well-known/pith/RFLE26KGE3X2VPU6ZC2YCUSAYW/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:RFLE26KGE3X2VPU6ZC2YCUSAYW","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":"8662b9b659273fab50a8d828a2af48bcce418a7600e221c21cc7d2f57fe6d887","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-27T06:24:53Z","title_canon_sha256":"b884c31331805090c034cbcdb47c36a4919a9804050dbc77c390daef8d08b61b"},"schema_version":"1.0","source":{"id":"1806.10317","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1806.10317","created_at":"2026-05-18T00:12:12Z"},{"alias_kind":"arxiv_version","alias_value":"1806.10317v1","created_at":"2026-05-18T00:12:12Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1806.10317","created_at":"2026-05-18T00:12:12Z"},{"alias_kind":"pith_short_12","alias_value":"RFLE26KGE3X2","created_at":"2026-05-18T12:32:50Z"},{"alias_kind":"pith_short_16","alias_value":"RFLE26KGE3X2VPU6","created_at":"2026-05-18T12:32:50Z"},{"alias_kind":"pith_short_8","alias_value":"RFLE26KG","created_at":"2026-05-18T12:32:50Z"}],"graph_snapshots":[{"event_id":"sha256:14bd3d584742b389ce510a63d2403869f9035fdd6d76e96046c697f7e43cf9d8","target":"graph","created_at":"2026-05-18T00:12:12Z","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":"Bayesian neural networks (BNNs) allow us to reason about uncertainty in a principled way. Stochastic Gradient Langevin Dynamics (SGLD) enables efficient BNN learning by drawing samples from the BNN posterior using mini-batches. However, SGLD and its extensions require storage of many copies of the model parameters, a potentially prohibitive cost, especially for large neural networks. We propose a framework, Adversarial Posterior Distillation, to distill the SGLD samples using a Generative Adversarial Network (GAN). At test-time, samples are generated by the GAN. We show that this distillation ","authors_text":"James Lucas, Kuan-Chieh Wang, Li Gu, Paul Vicol, Richard Zemel, Roger Grosse","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-27T06:24:53Z","title":"Adversarial Distillation of Bayesian Neural Network Posteriors"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1806.10317","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":"sha256:9ec5f2e35dbb819518b3bb3baeaf3558ded6ab4a35c722018df6e7c737d060be","target":"record","created_at":"2026-05-18T00:12:12Z","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":"8662b9b659273fab50a8d828a2af48bcce418a7600e221c21cc7d2f57fe6d887","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2018-06-27T06:24:53Z","title_canon_sha256":"b884c31331805090c034cbcdb47c36a4919a9804050dbc77c390daef8d08b61b"},"schema_version":"1.0","source":{"id":"1806.10317","kind":"arxiv","version":1}},"canonical_sha256":"89564d794626efaabe9ec8b5815240c58c6bf01eeb9807d389fd30984d8a1e2a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"89564d794626efaabe9ec8b5815240c58c6bf01eeb9807d389fd30984d8a1e2a","first_computed_at":"2026-05-18T00:12:12.834623Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:12:12.834623Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"kQQ0SASz7BEfml9H50N8vysUNwLYljI155turMriWofzvF6wmNGo8LHQ8LQE9PTfejIIRd2p0vg0IKZ9h0atBw==","signature_status":"signed_v1","signed_at":"2026-05-18T00:12:12.835354Z","signed_message":"canonical_sha256_bytes"},"source_id":"1806.10317","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:9ec5f2e35dbb819518b3bb3baeaf3558ded6ab4a35c722018df6e7c737d060be","sha256:14bd3d584742b389ce510a63d2403869f9035fdd6d76e96046c697f7e43cf9d8"],"state_sha256":"aec87fa0afd9a5b98b3be43d42e82cc5e42fa26a3ad98712e3e5e9e8ee2ac6f8"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"DhoPtV99OfCdpDFtjSxXHEtHLwl82ND20a3TmaSVG4o91aL21LpFkYlgYK6qZLrAZ3YTTPysV+8AG65ft+PDCw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-31T21:23:05.031950Z","bundle_sha256":"5104f97f6861ffef341e8a898ca3256fd414e8b727a875cf13566a21dffc3ca3"}}