{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:YWGBSHP3MGPG7TEGZIQP2XQDOG","short_pith_number":"pith:YWGBSHP3","canonical_record":{"source":{"id":"2107.05134","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-11T21:43:18Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"16a5756b80f91ed792b86ff974e964eb723c23a8a57fd9d52cf3e6295e31fbc4","abstract_canon_sha256":"1183ec60253c39627e746524c43346d7f92c06233b93bc5e8c0c4b419c961d30"},"schema_version":"1.0"},"canonical_sha256":"c58c191dfb619e6fcc86ca20fd5e0371a83d1829f9a2b2ee6aeac0b31f79ee2a","source":{"kind":"arxiv","id":"2107.05134","version":2},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.05134","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"arxiv_version","alias_value":"2107.05134v2","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.05134","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"pith_short_12","alias_value":"YWGBSHP3MGPG","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"pith_short_16","alias_value":"YWGBSHP3MGPG7TEG","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"pith_short_8","alias_value":"YWGBSHP3","created_at":"2026-07-05T03:56:47Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:YWGBSHP3MGPG7TEGZIQP2XQDOG","target":"record","payload":{"canonical_record":{"source":{"id":"2107.05134","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-11T21:43:18Z","cross_cats_sorted":["math.OC","stat.ML"],"title_canon_sha256":"16a5756b80f91ed792b86ff974e964eb723c23a8a57fd9d52cf3e6295e31fbc4","abstract_canon_sha256":"1183ec60253c39627e746524c43346d7f92c06233b93bc5e8c0c4b419c961d30"},"schema_version":"1.0"},"canonical_sha256":"c58c191dfb619e6fcc86ca20fd5e0371a83d1829f9a2b2ee6aeac0b31f79ee2a","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T03:56:47.839874Z","signature_b64":"FapY8ZWqPfKviYRqG8HZUwTxIE4QnJpiA9u6WqtQZmm0iMELSKn/zztEE7YgTVg+bb4kax9ytWi7YTmeWLl7Bg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"c58c191dfb619e6fcc86ca20fd5e0371a83d1829f9a2b2ee6aeac0b31f79ee2a","last_reissued_at":"2026-07-05T03:56:47.839386Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T03:56:47.839386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2107.05134","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":"2026-07-05T03:56:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"ubpzf4ZzbJql1hLHCeIV3x6U3Wujh2RWwZLme9SQdb+/m9rrE00o7M0ehWXT9KRFtQDhyzJBk2ASGavHg423AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:13:20.326121Z"},"content_sha256":"caaefadcfb71791ef97b6d49d9bf6f138f2778fc1ef4e476db5d2c17e9161bf4","schema_version":"1.0","event_id":"sha256:caaefadcfb71791ef97b6d49d9bf6f138f2778fc1ef4e476db5d2c17e9161bf4"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:YWGBSHP3MGPG7TEGZIQP2XQDOG","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"Dual Training of Energy-Based Models with Overparametrized Shallow Neural Networks","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["math.OC","stat.ML"],"primary_cat":"cs.LG","authors_text":"Alberto Bietti, Carles Domingo-Enrich, Eric Vanden-Eijnden, Joan Bruna, Marylou Gabri\\'e","submitted_at":"2021-07-11T21:43:18Z","abstract_excerpt":"Energy-based models (EBMs) are generative models that are usually trained via maximum likelihood estimation. This approach becomes challenging in generic situations where the trained energy is non-convex, due to the need to sample the Gibbs distribution associated with this energy. Using general Fenchel duality results, we derive variational principles dual to maximum likelihood EBMs with shallow overparametrized neural network energies, both in the feature-learning and lazy linearized regimes. In the feature-learning regime, this dual formulation justifies using a two time-scale gradient asce"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.05134","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":""},"integrity":{"clean":true,"summary":{"advisory":0,"critical":0,"by_detector":{},"informational":0},"endpoint":"/pith/2107.05134/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":"2026-07-05T03:56:47Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"HZK+m2x5v6MY9tSSx9xe3HM3P47MQK8pbT/Op4f+q8Dw8X8DuCTP0AbdE16vyxv/xp/Hr9f6W2i9+XyqJt78AQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-08T13:13:20.326734Z"},"content_sha256":"b6f642abb2935dbfbc23778366e504a2d7f58b4e736de29e4364e2abb49377e7","schema_version":"1.0","event_id":"sha256:b6f642abb2935dbfbc23778366e504a2d7f58b4e736de29e4364e2abb49377e7"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/YWGBSHP3MGPG7TEGZIQP2XQDOG/bundle.json","state_url":"https://pith.science/pith/YWGBSHP3MGPG7TEGZIQP2XQDOG/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/YWGBSHP3MGPG7TEGZIQP2XQDOG/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":"2026-08-08T13:13:20Z","links":{"resolver":"https://pith.science/pith/YWGBSHP3MGPG7TEGZIQP2XQDOG","bundle":"https://pith.science/pith/YWGBSHP3MGPG7TEGZIQP2XQDOG/bundle.json","state":"https://pith.science/pith/YWGBSHP3MGPG7TEGZIQP2XQDOG/state.json","well_known_bundle":"https://pith.science/.well-known/pith/YWGBSHP3MGPG7TEGZIQP2XQDOG/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:YWGBSHP3MGPG7TEGZIQP2XQDOG","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":"1183ec60253c39627e746524c43346d7f92c06233b93bc5e8c0c4b419c961d30","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-11T21:43:18Z","title_canon_sha256":"16a5756b80f91ed792b86ff974e964eb723c23a8a57fd9d52cf3e6295e31fbc4"},"schema_version":"1.0","source":{"id":"2107.05134","kind":"arxiv","version":2}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2107.05134","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"arxiv_version","alias_value":"2107.05134v2","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2107.05134","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"pith_short_12","alias_value":"YWGBSHP3MGPG","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"pith_short_16","alias_value":"YWGBSHP3MGPG7TEG","created_at":"2026-07-05T03:56:47Z"},{"alias_kind":"pith_short_8","alias_value":"YWGBSHP3","created_at":"2026-07-05T03:56:47Z"}],"graph_snapshots":[{"event_id":"sha256:b6f642abb2935dbfbc23778366e504a2d7f58b4e736de29e4364e2abb49377e7","target":"graph","created_at":"2026-07-05T03:56:47Z","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/2107.05134/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"Energy-based models (EBMs) are generative models that are usually trained via maximum likelihood estimation. This approach becomes challenging in generic situations where the trained energy is non-convex, due to the need to sample the Gibbs distribution associated with this energy. Using general Fenchel duality results, we derive variational principles dual to maximum likelihood EBMs with shallow overparametrized neural network energies, both in the feature-learning and lazy linearized regimes. In the feature-learning regime, this dual formulation justifies using a two time-scale gradient asce","authors_text":"Alberto Bietti, Carles Domingo-Enrich, Eric Vanden-Eijnden, Joan Bruna, Marylou Gabri\\'e","cross_cats":["math.OC","stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-11T21:43:18Z","title":"Dual Training of Energy-Based Models with Overparametrized Shallow Neural Networks"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2107.05134","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:caaefadcfb71791ef97b6d49d9bf6f138f2778fc1ef4e476db5d2c17e9161bf4","target":"record","created_at":"2026-07-05T03:56:47Z","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":"1183ec60253c39627e746524c43346d7f92c06233b93bc5e8c0c4b419c961d30","cross_cats_sorted":["math.OC","stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-07-11T21:43:18Z","title_canon_sha256":"16a5756b80f91ed792b86ff974e964eb723c23a8a57fd9d52cf3e6295e31fbc4"},"schema_version":"1.0","source":{"id":"2107.05134","kind":"arxiv","version":2}},"canonical_sha256":"c58c191dfb619e6fcc86ca20fd5e0371a83d1829f9a2b2ee6aeac0b31f79ee2a","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"c58c191dfb619e6fcc86ca20fd5e0371a83d1829f9a2b2ee6aeac0b31f79ee2a","first_computed_at":"2026-07-05T03:56:47.839386Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T03:56:47.839386Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"FapY8ZWqPfKviYRqG8HZUwTxIE4QnJpiA9u6WqtQZmm0iMELSKn/zztEE7YgTVg+bb4kax9ytWi7YTmeWLl7Bg==","signature_status":"signed_v1","signed_at":"2026-07-05T03:56:47.839874Z","signed_message":"canonical_sha256_bytes"},"source_id":"2107.05134","source_kind":"arxiv","source_version":2}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:caaefadcfb71791ef97b6d49d9bf6f138f2778fc1ef4e476db5d2c17e9161bf4","sha256:b6f642abb2935dbfbc23778366e504a2d7f58b4e736de29e4364e2abb49377e7"],"state_sha256":"ad5dc9a626d9f46bda1caee7a938999ea811920bb7191fd1a333ea525d3b6494"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"A7unZ4/3UKu9Ztdqp7qXmIfHgB6BS+VuE0CNz2uUclZ9lu2JWBAW+tUzUsKMXiaI4guS3yKFse21yyYVgFB/Ag==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-08T13:13:20.330715Z","bundle_sha256":"ab19bde1efe03cd70fe1cb66f3df5040a8a7ba4d729efd42da2c82de1e13e8d0"}}