{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2018:WZV3EY7QNSVU57W4IGC4BOIFBT","short_pith_number":"pith:WZV3EY7Q","canonical_record":{"source":{"id":"1811.08725","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-11-21T13:41:21Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"031825fdaf1b36dcf59c4bec591b272f755b6966e1a0892da4e49b15e2bd6904","abstract_canon_sha256":"a31179169593b9878134d85fe21ce8306ab041248fd5e726e70eb51c335c4ac0"},"schema_version":"1.0"},"canonical_sha256":"b66bb263f06cab4efedc4185c0b9050cec9b083c4df1978e82891298da5b99a4","source":{"kind":"arxiv","id":"1811.08725","version":1},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.08725","created_at":"2026-05-18T00:00:11Z"},{"alias_kind":"arxiv_version","alias_value":"1811.08725v1","created_at":"2026-05-18T00:00:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.08725","created_at":"2026-05-18T00:00:11Z"},{"alias_kind":"pith_short_12","alias_value":"WZV3EY7QNSVU","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_16","alias_value":"WZV3EY7QNSVU57W4","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_8","alias_value":"WZV3EY7Q","created_at":"2026-05-18T12:33:01Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2018:WZV3EY7QNSVU57W4IGC4BOIFBT","target":"record","payload":{"canonical_record":{"source":{"id":"1811.08725","kind":"arxiv","version":1},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-11-21T13:41:21Z","cross_cats_sorted":["cs.LG"],"title_canon_sha256":"031825fdaf1b36dcf59c4bec591b272f755b6966e1a0892da4e49b15e2bd6904","abstract_canon_sha256":"a31179169593b9878134d85fe21ce8306ab041248fd5e726e70eb51c335c4ac0"},"schema_version":"1.0"},"canonical_sha256":"b66bb263f06cab4efedc4185c0b9050cec9b083c4df1978e82891298da5b99a4","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-05-18T00:00:11.766689Z","signature_b64":"9Iw82KFik2EdO41X0UohXMrY6kod5PoxNz/v2bJ58+M/+B6TksoSYPRgr4faYAlN9/2FkfnbIk8u4fYFIbULAQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"b66bb263f06cab4efedc4185c0b9050cec9b083c4df1978e82891298da5b99a4","last_reissued_at":"2026-05-18T00:00:11.766071Z","signature_status":"signed_v1","first_computed_at":"2026-05-18T00:00:11.766071Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"1811.08725","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":"2026-05-18T00:00:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"inRZd/2IHRrQJhwaURJVOJ12Xs9IqNhPaeBq5mBQxuwp+lDS4CdzfCW8kCtwtAdbYCrMEdULD64UemtRfFtbBQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T06:58:32.413887Z"},"content_sha256":"427c55395dd001f1b5ca3e3a68faf6833ad3a22db1606744b1a7cc94dc2e6d41","schema_version":"1.0","event_id":"sha256:427c55395dd001f1b5ca3e3a68faf6833ad3a22db1606744b1a7cc94dc2e6d41"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2018:WZV3EY7QNSVU57W4IGC4BOIFBT","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Marginal Weighted Maximum Log-likelihood for Efficient Learning of Perturb-and-Map models","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.LG"],"primary_cat":"stat.ML","authors_text":"Anton Osokin, Francis Bach, Tatiana Shpakova","submitted_at":"2018-11-21T13:41:21Z","abstract_excerpt":"We consider the structured-output prediction problem through probabilistic approaches and generalize the \"perturb-and-MAP\" framework to more challenging weighted Hamming losses, which are crucial in applications. While in principle our approach is a straightforward marginalization, it requires solving many related MAP inference problems. We show that for log-supermodular pairwise models these operations can be performed efficiently using the machinery of dynamic graph cuts. We also propose to use double stochastic gradient descent, both on the data and on the perturbations, for efficient learn"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.08725","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":"2026-05-18T00:00:11Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"201p0It57irwkw5PPWWKzG4oP+pO62Xqp//tjukulPxs1PAboWkKt9kf3ewZBySSAUGUEDT8ybOgh6PUpJfWAA==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-05-30T06:58:32.414558Z"},"content_sha256":"9730e91410d98879ed37052a23be63f9f336b2c4f40af2ffb0447c4fd96b82e9","schema_version":"1.0","event_id":"sha256:9730e91410d98879ed37052a23be63f9f336b2c4f40af2ffb0447c4fd96b82e9"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/WZV3EY7QNSVU57W4IGC4BOIFBT/bundle.json","state_url":"https://pith.science/pith/WZV3EY7QNSVU57W4IGC4BOIFBT/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/WZV3EY7QNSVU57W4IGC4BOIFBT/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":"2026-05-30T06:58:32Z","links":{"resolver":"https://pith.science/pith/WZV3EY7QNSVU57W4IGC4BOIFBT","bundle":"https://pith.science/pith/WZV3EY7QNSVU57W4IGC4BOIFBT/bundle.json","state":"https://pith.science/pith/WZV3EY7QNSVU57W4IGC4BOIFBT/state.json","well_known_bundle":"https://pith.science/.well-known/pith/WZV3EY7QNSVU57W4IGC4BOIFBT/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2018:WZV3EY7QNSVU57W4IGC4BOIFBT","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":"a31179169593b9878134d85fe21ce8306ab041248fd5e726e70eb51c335c4ac0","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-11-21T13:41:21Z","title_canon_sha256":"031825fdaf1b36dcf59c4bec591b272f755b6966e1a0892da4e49b15e2bd6904"},"schema_version":"1.0","source":{"id":"1811.08725","kind":"arxiv","version":1}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"1811.08725","created_at":"2026-05-18T00:00:11Z"},{"alias_kind":"arxiv_version","alias_value":"1811.08725v1","created_at":"2026-05-18T00:00:11Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.1811.08725","created_at":"2026-05-18T00:00:11Z"},{"alias_kind":"pith_short_12","alias_value":"WZV3EY7QNSVU","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_16","alias_value":"WZV3EY7QNSVU57W4","created_at":"2026-05-18T12:33:01Z"},{"alias_kind":"pith_short_8","alias_value":"WZV3EY7Q","created_at":"2026-05-18T12:33:01Z"}],"graph_snapshots":[{"event_id":"sha256:9730e91410d98879ed37052a23be63f9f336b2c4f40af2ffb0447c4fd96b82e9","target":"graph","created_at":"2026-05-18T00:00:11Z","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":"We consider the structured-output prediction problem through probabilistic approaches and generalize the \"perturb-and-MAP\" framework to more challenging weighted Hamming losses, which are crucial in applications. While in principle our approach is a straightforward marginalization, it requires solving many related MAP inference problems. We show that for log-supermodular pairwise models these operations can be performed efficiently using the machinery of dynamic graph cuts. We also propose to use double stochastic gradient descent, both on the data and on the perturbations, for efficient learn","authors_text":"Anton Osokin, Francis Bach, Tatiana Shpakova","cross_cats":["cs.LG"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-11-21T13:41:21Z","title":"Marginal Weighted Maximum Log-likelihood for Efficient Learning of Perturb-and-Map models"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"1811.08725","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":"sha256:427c55395dd001f1b5ca3e3a68faf6833ad3a22db1606744b1a7cc94dc2e6d41","target":"record","created_at":"2026-05-18T00:00:11Z","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":"a31179169593b9878134d85fe21ce8306ab041248fd5e726e70eb51c335c4ac0","cross_cats_sorted":["cs.LG"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"stat.ML","submitted_at":"2018-11-21T13:41:21Z","title_canon_sha256":"031825fdaf1b36dcf59c4bec591b272f755b6966e1a0892da4e49b15e2bd6904"},"schema_version":"1.0","source":{"id":"1811.08725","kind":"arxiv","version":1}},"canonical_sha256":"b66bb263f06cab4efedc4185c0b9050cec9b083c4df1978e82891298da5b99a4","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"b66bb263f06cab4efedc4185c0b9050cec9b083c4df1978e82891298da5b99a4","first_computed_at":"2026-05-18T00:00:11.766071Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-05-18T00:00:11.766071Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"9Iw82KFik2EdO41X0UohXMrY6kod5PoxNz/v2bJ58+M/+B6TksoSYPRgr4faYAlN9/2FkfnbIk8u4fYFIbULAQ==","signature_status":"signed_v1","signed_at":"2026-05-18T00:00:11.766689Z","signed_message":"canonical_sha256_bytes"},"source_id":"1811.08725","source_kind":"arxiv","source_version":1}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:427c55395dd001f1b5ca3e3a68faf6833ad3a22db1606744b1a7cc94dc2e6d41","sha256:9730e91410d98879ed37052a23be63f9f336b2c4f40af2ffb0447c4fd96b82e9"],"state_sha256":"ecd7af9794716c38c9cd47374767afe8199f993b2fc395dd0f46b1ead057e629"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"s9ca96HP2GAAnfP0q7YlIeqRw5bKeFjajIe2lGv2wwwFhw4WZ2hx+1kYLuDVAa5P8cmC18c9KUOW9yfakUy/CQ==","signed_message":"bundle_sha256_bytes","signed_at":"2026-05-30T06:58:32.417062Z","bundle_sha256":"005da7d684652cb02460b1342e977c2dac7d1d769cb2641223061636a2eca877"}}