{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:2Q3HFB2O6FEU5XS6FWCIYEMO3H","short_pith_number":"pith:2Q3HFB2O","canonical_record":{"source":{"id":"1805.08235","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-05-21T18:06:53Z","cross_cats_sorted":[],"title_canon_sha256":"4f15d862a203f8fe03ba0b977becdbb9c50d87ff7fda0d295e8d2b8a1f3a163a","abstract_canon_sha256":"f3b0d38901d91111e3fa0dedf11b7c8eb9af86d02528cca09f073a71ed5c89fc"},"schema_version":"1.0"},"canonical_sha256":"d43672874ef1494ede5e2d848c118ed9eb62ca63f883a7f0dd3810c54448667b","source":{"kind":"arxiv","id":"1805.08235","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.08235","created_at":"2026-05-17T23:49:05Z"},{"alias_kind":"arxiv_version","alias_value":"1805.08235v2","created_at":"2026-05-17T23:49:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.08235","created_at":"2026-05-17T23:49:05Z"},{"alias_kind":"pith_short_12","alias_value":"2Q3HFB2O6FEU","created_at":"2026-05-18T12:32:02Z"},{"alias_kind":"pith_short_16","alias_value":"2Q3HFB2O6FEU5XS6","created_at":"2026-05-18T12:32:02Z"},{"alias_kind":"pith_short_8","alias_value":"2Q3HFB2O","created_at":"2026-05-18T12:32:02Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:2Q3HFB2O6FEU5XS6FWCIYEMO3H","target":"record","payload":{"canonical_record":{"source":{"id":"1805.08235","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-05-21T18:06:53Z","cross_cats_sorted":[],"title_canon_sha256":"4f15d862a203f8fe03ba0b977becdbb9c50d87ff7fda0d295e8d2b8a1f3a163a","abstract_canon_sha256":"f3b0d38901d91111e3fa0dedf11b7c8eb9af86d02528cca09f073a71ed5c89fc"},"schema_version":"1.0"},"canonical_sha256":"d43672874ef1494ede5e2d848c118ed9eb62ca63f883a7f0dd3810c54448667b","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-17T23:49:05.679707Z","signature_b64":"fA4n5JayfVC//lFfaGOISDV+lCGCePqX6QEBYQGWSgK8fVFMP/XmywgCtKekR4comj4F6fuCbWp2q3ig+I6DAw==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"d43672874ef1494ede5e2d848c118ed9eb62ca63f883a7f0dd3810c54448667b","last_reissued_at":"2026-05-17T23:49:05.678907Z","signature_status":"signed_v1","first_computed_at":"2026-05-17T23:49:05.678907Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1805.08235","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":"2026-05-17T23:49:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"rVyGw50kfsgwVt0dSRgFv6TCGzz2diOL9UBipm/FidhxuaHY1mb/gMD5OU/jUyr7WIsYdkTE+8tGdG6L0cFcBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-12T03:38:55.403207Z"},"content_sha256":"c560bd6b870174a9caa47f965ed7c819c862d58e09edb59b7fa33b626e2afdb2","schema_version":"1.0","event_id":"sha256:c560bd6b870174a9caa47f965ed7c819c862d58e09edb59b7fa33b626e2afdb2"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:2Q3HFB2O6FEU5XS6FWCIYEMO3H","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Improving CNN classifiers by estimating test-time priors","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":[],"primary_cat":"cs.CV","authors_text":"Jiri Matas, Milan Sulc","submitted_at":"2018-05-21T18:06:53Z","abstract_excerpt":"The problem of different training and test set class priors is addressed in the context of CNN classifiers. We compare two different approaches to estimating the new priors: an existing Maximum Likelihood Estimation approach (optimized by an EM algorithm or by projected gradient descend) and a proposed Maximum a Posteriori approach, which increases the stability of the estimate by introducing a Dirichlet hyper-prior on the class prior probabilities. Experimental results show a significant improvement on the fine-grained classification tasks using known evaluation-time priors, increasing the to"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.08235","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":"2026-05-17T23:49:05Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"6o4pnJi4jCv11ejeLPuEZ2GP4zXgbU68OZYv2RjRY7ox2NKzE05PMoiiE4MUm3uu1kS5ZNNLEtfRtEYUb8KIDw==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-06-12T03:38:55.403857Z"},"content_sha256":"6bcd0dbf39bc839d631d3ff6988ee008a3588d8d7f9e787e5cecd45b409160c7","schema_version":"1.0","event_id":"sha256:6bcd0dbf39bc839d631d3ff6988ee008a3588d8d7f9e787e5cecd45b409160c7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/2Q3HFB2O6FEU5XS6FWCIYEMO3H/bundle.json","state_url":"https://pith.science/pith/2Q3HFB2O6FEU5XS6FWCIYEMO3H/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/2Q3HFB2O6FEU5XS6FWCIYEMO3H/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":"2026-06-12T03:38:55Z","links":{"resolver":"https://pith.science/pith/2Q3HFB2O6FEU5XS6FWCIYEMO3H","bundle":"https://pith.science/pith/2Q3HFB2O6FEU5XS6FWCIYEMO3H/bundle.json","state":"https://pith.science/pith/2Q3HFB2O6FEU5XS6FWCIYEMO3H/state.json","well_known_bundle":"https://pith.science/.well-known/pith/2Q3HFB2O6FEU5XS6FWCIYEMO3H/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:2Q3HFB2O6FEU5XS6FWCIYEMO3H","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":"f3b0d38901d91111e3fa0dedf11b7c8eb9af86d02528cca09f073a71ed5c89fc","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-05-21T18:06:53Z","title_canon_sha256":"4f15d862a203f8fe03ba0b977becdbb9c50d87ff7fda0d295e8d2b8a1f3a163a"},"schema_version":"1.0","source":{"id":"1805.08235","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1805.08235","created_at":"2026-05-17T23:49:05Z"},{"alias_kind":"arxiv_version","alias_value":"1805.08235v2","created_at":"2026-05-17T23:49:05Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1805.08235","created_at":"2026-05-17T23:49:05Z"},{"alias_kind":"pith_short_12","alias_value":"2Q3HFB2O6FEU","created_at":"2026-05-18T12:32:02Z"},{"alias_kind":"pith_short_16","alias_value":"2Q3HFB2O6FEU5XS6","created_at":"2026-05-18T12:32:02Z"},{"alias_kind":"pith_short_8","alias_value":"2Q3HFB2O","created_at":"2026-05-18T12:32:02Z"}],"graph_snapshots":[{"event_id":"sha256:6bcd0dbf39bc839d631d3ff6988ee008a3588d8d7f9e787e5cecd45b409160c7","target":"graph","created_at":"2026-05-17T23:49:05Z","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":"The problem of different training and test set class priors is addressed in the context of CNN classifiers. We compare two different approaches to estimating the new priors: an existing Maximum Likelihood Estimation approach (optimized by an EM algorithm or by projected gradient descend) and a proposed Maximum a Posteriori approach, which increases the stability of the estimate by introducing a Dirichlet hyper-prior on the class prior probabilities. Experimental results show a significant improvement on the fine-grained classification tasks using known evaluation-time priors, increasing the to","authors_text":"Jiri Matas, Milan Sulc","cross_cats":[],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-05-21T18:06:53Z","title":"Improving CNN classifiers by estimating test-time priors"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1805.08235","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:c560bd6b870174a9caa47f965ed7c819c862d58e09edb59b7fa33b626e2afdb2","target":"record","created_at":"2026-05-17T23:49:05Z","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":"f3b0d38901d91111e3fa0dedf11b7c8eb9af86d02528cca09f073a71ed5c89fc","cross_cats_sorted":[],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.CV","submitted_at":"2018-05-21T18:06:53Z","title_canon_sha256":"4f15d862a203f8fe03ba0b977becdbb9c50d87ff7fda0d295e8d2b8a1f3a163a"},"schema_version":"1.0","source":{"id":"1805.08235","kind":"arxiv","version":2}},"canonical_sha256":"d43672874ef1494ede5e2d848c118ed9eb62ca63f883a7f0dd3810c54448667b","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"d43672874ef1494ede5e2d848c118ed9eb62ca63f883a7f0dd3810c54448667b","first_computed_at":"2026-05-17T23:49:05.678907Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-17T23:49:05.678907Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"fA4n5JayfVC//lFfaGOISDV+lCGCePqX6QEBYQGWSgK8fVFMP/XmywgCtKekR4comj4F6fuCbWp2q3ig+I6DAw==","signature_status":"signed_v1","signed_at":"2026-05-17T23:49:05.679707Z","signed_message":"canonical_sha256_bytes"},"source_id":"1805.08235","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:c560bd6b870174a9caa47f965ed7c819c862d58e09edb59b7fa33b626e2afdb2","sha256:6bcd0dbf39bc839d631d3ff6988ee008a3588d8d7f9e787e5cecd45b409160c7"],"state_sha256":"9289bbe2eeca62e4a75e2593039b550bf09b843c55de9aeebb90b2a714dafeff"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"VUf7TAOaVYnyZ9mHKWXQBVUSb3kxF/HdHOiSw6QO68DQw0TM+yQkJnVc6USOiE3SW7KWuX5iSJfyhyeLAaBMDA==","signed_message":"bundle_sha256_bytes","signed_at":"2026-06-12T03:38:55.407220Z","bundle_sha256":"766ff36afed6dab9164015555f7f98563c66130e37f381df5e98e5857e0f38ea"}}