{"bundle_type":"pith_open_graph_bundle","bundle_version":"1.0","pith_number":"pith:2021:VJTSK4VDQZN7SO6HAJRZ53FWNO","short_pith_number":"pith:VJTSK4VD","canonical_record":{"source":{"id":"2112.15577","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-12-31T18:03:46Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"545950279c1d3ef578b6c041b992cf2cced68997817d1b0747247701ce2f77de","abstract_canon_sha256":"d535bb6d303a8b79f9878098842844d6e8b796d8606de0fa6c2b5b9c963a7e7d"},"schema_version":"1.0"},"canonical_sha256":"aa672572a3865bf93bc702639eecb66b981d9643edac727d8216f7670695d999","source":{"kind":"arxiv","id":"2112.15577","version":4},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.15577","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"arxiv_version","alias_value":"2112.15577v4","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.15577","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"pith_short_12","alias_value":"VJTSK4VDQZN7","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"pith_short_16","alias_value":"VJTSK4VDQZN7SO6H","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"pith_short_8","alias_value":"VJTSK4VD","created_at":"2026-07-05T05:08:25Z"}],"events":[{"event_type":"record_created","subject_pith_number":"pith:2021:VJTSK4VDQZN7SO6HAJRZ53FWNO","target":"record","payload":{"canonical_record":{"source":{"id":"2112.15577","kind":"arxiv","version":4},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-12-31T18:03:46Z","cross_cats_sorted":["stat.ML"],"title_canon_sha256":"545950279c1d3ef578b6c041b992cf2cced68997817d1b0747247701ce2f77de","abstract_canon_sha256":"d535bb6d303a8b79f9878098842844d6e8b796d8606de0fa6c2b5b9c963a7e7d"},"schema_version":"1.0"},"canonical_sha256":"aa672572a3865bf93bc702639eecb66b981d9643edac727d8216f7670695d999","receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T05:08:25.890740Z","signature_b64":"Z4utTXtqw7K+6Vjpob+z/5Rf80m0X6YM1z6kAogEVPjwnS8t84HFrDT/l/ntWJhcx/9pa80mUNWagO+u37SyDg==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"aa672572a3865bf93bc702639eecb66b981d9643edac727d8216f7670695d999","last_reissued_at":"2026-07-05T05:08:25.890344Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T05:08:25.890344Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"source_kind":"arxiv","source_id":"2112.15577","source_version":4,"attestation_state":"computed"},"signer":{"signer_id":"pith.science","signer_type":"pith_registry","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"created_at":"2026-07-05T05:08:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"2IRPEWcPq3DLa7vLL435CVMkWw6UV7/ECW50VICeedJn9jLgWYfiYyqk1K9mdCNQE75pd3wExCf1ODWDsX5ZAQ==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T22:35:23.521018Z"},"content_sha256":"269cb139e4c467bfb0a445bca8f03dfa03a4ac03c9e6294f5e0ecac18180759b","schema_version":"1.0","event_id":"sha256:269cb139e4c467bfb0a445bca8f03dfa03a4ac03c9e6294f5e0ecac18180759b"},{"event_type":"graph_snapshot","subject_pith_number":"pith:2021:VJTSK4VDQZN7SO6HAJRZ53FWNO","target":"graph","payload":{"graph_snapshot":{"paper":{"title":"How Infinitely Wide Neural Networks Can Benefit from Multi-task Learning -- an Exact Macroscopic Characterization","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["stat.ML"],"primary_cat":"cs.LG","authors_text":"Hanna Wutte, Jakob Heiss, Josef Teichmann","submitted_at":"2021-12-31T18:03:46Z","abstract_excerpt":"In practice, multi-task learning (through learning features shared among tasks) is an essential property of deep neural networks (NNs). While infinite-width limits of NNs can provide good intuition for their generalization behavior, the well-known infinite-width limits of NNs in the literature (e.g., neural tangent kernels) assume specific settings in which wide ReLU-NNs behave like shallow Gaussian Processes with a fixed kernel. Consequently, in such settings, these NNs lose their ability to benefit from multi-task learning in the infinite-width limit. In contrast, we prove that optimizing wi"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.15577","kind":"arxiv","version":4},"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/2112.15577/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":"2026-07-05T05:08:25Z","supersedes":[],"prev_event":null,"signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"w5Ngv3iw+TV3dtv24De/J5j3fNhc7zc1jk9LoLZQ/RekLKVcn5TD9gfBpZlcKSgAYRGJMGVBHv58Y/DQvwNjBg==","signed_message":"open_graph_event_sha256_bytes","signed_at":"2026-08-12T22:35:23.521631Z"},"content_sha256":"17f0e384e87a28dfacd5b610e1a2d642a06626bc9d78c9120e5e555344e490d3","schema_version":"1.0","event_id":"sha256:17f0e384e87a28dfacd5b610e1a2d642a06626bc9d78c9120e5e555344e490d3"}],"timestamp_proofs":[],"mirror_hints":[{"mirror_type":"https","name":"Pith Resolver","base_url":"https://pith.science","bundle_url":"https://pith.science/pith/VJTSK4VDQZN7SO6HAJRZ53FWNO/bundle.json","state_url":"https://pith.science/pith/VJTSK4VDQZN7SO6HAJRZ53FWNO/state.json","well_known_bundle_url":"https://pith.science/.well-known/pith/VJTSK4VDQZN7SO6HAJRZ53FWNO/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":"2026-08-12T22:35:23Z","links":{"resolver":"https://pith.science/pith/VJTSK4VDQZN7SO6HAJRZ53FWNO","bundle":"https://pith.science/pith/VJTSK4VDQZN7SO6HAJRZ53FWNO/bundle.json","state":"https://pith.science/pith/VJTSK4VDQZN7SO6HAJRZ53FWNO/state.json","well_known_bundle":"https://pith.science/.well-known/pith/VJTSK4VDQZN7SO6HAJRZ53FWNO/bundle.json"},"state":{"state_type":"pith_open_graph_state","state_version":"1.0","pith_number":"pith:2021:VJTSK4VDQZN7SO6HAJRZ53FWNO","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":"d535bb6d303a8b79f9878098842844d6e8b796d8606de0fa6c2b5b9c963a7e7d","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-12-31T18:03:46Z","title_canon_sha256":"545950279c1d3ef578b6c041b992cf2cced68997817d1b0747247701ce2f77de"},"schema_version":"1.0","source":{"id":"2112.15577","kind":"arxiv","version":4}},"source_aliases":[{"alias_kind":"arxiv","alias_value":"2112.15577","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"arxiv_version","alias_value":"2112.15577v4","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2112.15577","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"pith_short_12","alias_value":"VJTSK4VDQZN7","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"pith_short_16","alias_value":"VJTSK4VDQZN7SO6H","created_at":"2026-07-05T05:08:25Z"},{"alias_kind":"pith_short_8","alias_value":"VJTSK4VD","created_at":"2026-07-05T05:08:25Z"}],"graph_snapshots":[{"event_id":"sha256:17f0e384e87a28dfacd5b610e1a2d642a06626bc9d78c9120e5e555344e490d3","target":"graph","created_at":"2026-07-05T05:08:25Z","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/2112.15577/integrity.json","findings":[],"snapshot_sha256":"c28c3603d3b5d939e8dc4c7e95fa8dfce3d595e45f758748cecf8e644a296938","summary":{"advisory":0,"by_detector":{},"critical":0,"informational":0}},"paper":{"abstract_excerpt":"In practice, multi-task learning (through learning features shared among tasks) is an essential property of deep neural networks (NNs). While infinite-width limits of NNs can provide good intuition for their generalization behavior, the well-known infinite-width limits of NNs in the literature (e.g., neural tangent kernels) assume specific settings in which wide ReLU-NNs behave like shallow Gaussian Processes with a fixed kernel. Consequently, in such settings, these NNs lose their ability to benefit from multi-task learning in the infinite-width limit. In contrast, we prove that optimizing wi","authors_text":"Hanna Wutte, Jakob Heiss, Josef Teichmann","cross_cats":["stat.ML"],"headline":"","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-12-31T18:03:46Z","title":"How Infinitely Wide Neural Networks Can Benefit from Multi-task Learning -- an Exact Macroscopic Characterization"},"references":{"count":0,"internal_anchors":0,"resolved_work":0,"sample":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2112.15577","kind":"arxiv","version":4},"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":"sha256:269cb139e4c467bfb0a445bca8f03dfa03a4ac03c9e6294f5e0ecac18180759b","target":"record","created_at":"2026-07-05T05:08:25Z","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":"d535bb6d303a8b79f9878098842844d6e8b796d8606de0fa6c2b5b9c963a7e7d","cross_cats_sorted":["stat.ML"],"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2021-12-31T18:03:46Z","title_canon_sha256":"545950279c1d3ef578b6c041b992cf2cced68997817d1b0747247701ce2f77de"},"schema_version":"1.0","source":{"id":"2112.15577","kind":"arxiv","version":4}},"canonical_sha256":"aa672572a3865bf93bc702639eecb66b981d9643edac727d8216f7670695d999","receipt":{"algorithm":"ed25519","builder_version":"pith-number-builder-2026-05-17-v1","canonical_sha256":"aa672572a3865bf93bc702639eecb66b981d9643edac727d8216f7670695d999","first_computed_at":"2026-07-05T05:08:25.890344Z","key_id":"pith-v1-2026-05","kind":"pith_receipt","last_reissued_at":"2026-07-05T05:08:25.890344Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","receipt_version":"0.3","signature_b64":"Z4utTXtqw7K+6Vjpob+z/5Rf80m0X6YM1z6kAogEVPjwnS8t84HFrDT/l/ntWJhcx/9pa80mUNWagO+u37SyDg==","signature_status":"signed_v1","signed_at":"2026-07-05T05:08:25.890740Z","signed_message":"canonical_sha256_bytes"},"source_id":"2112.15577","source_kind":"arxiv","source_version":4}}},"equivocations":[],"invalid_events":[],"applied_event_ids":["sha256:269cb139e4c467bfb0a445bca8f03dfa03a4ac03c9e6294f5e0ecac18180759b","sha256:17f0e384e87a28dfacd5b610e1a2d642a06626bc9d78c9120e5e555344e490d3"],"state_sha256":"5820d378b10c3f687440fc9588b0fc51d82d4547c101f5920782c0552c70f3b2"},"bundle_signature":{"signature_status":"signed_v1","algorithm":"ed25519","key_id":"pith-v1-2026-05","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54","signature_b64":"vbTyHeKmIkQUvWcSZmuwOBSiYkNkrL0n4gCGCwECgHdZ0MhBCyOI4YPzUV8ajoQzO4LEvkyxaRODYyzyaWaXAw==","signed_message":"bundle_sha256_bytes","signed_at":"2026-08-12T22:35:23.527072Z","bundle_sha256":"b51cf4a6877d933552904805845a95e924b5eea6baf63397080aeceac33dbb76"}}