{"record_type":"pith_number_record","schema_url":"https://pith.science/schemas/pith-number/v1.json","pith_number":"pith:2023:RDKOLTUXAYOAAJYXRUCMMKVR6O","short_pith_number":"pith:RDKOLTUX","schema_version":"1.0","canonical_sha256":"88d4e5ce97061c0027178d04c62ab1f38e19423d50f958bbf83e5f493e44f291","source":{"kind":"arxiv","id":"2306.04406","version":2},"attestation_state":"computed","paper":{"title":"Generalized Teacher Forcing for Learning Chaotic Dynamics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.DS","nlin.CD"],"primary_cat":"cs.LG","authors_text":"Daniel Durstewitz, Florian Hess, Manuel Brenner, Zahra Monfared","submitted_at":"2023-06-07T13:04:34Z","abstract_excerpt":"Chaotic dynamical systems (DS) are ubiquitous in nature and society. Often we are interested in reconstructing such systems from observed time series for prediction or mechanistic insight, where by reconstruction we mean learning geometrical and invariant temporal properties of the system in question (like attractors). However, training reconstruction algorithms like recurrent neural networks (RNNs) on such systems by gradient-descent based techniques faces severe challenges. This is mainly due to exploding gradients caused by the exponential divergence of trajectories in chaotic systems. More"},"verification_status":{"content_addressed":true,"pith_receipt":true,"author_attested":false,"weak_author_claims":0,"strong_author_claims":0,"externally_anchored":false,"storage_verified":false,"citation_signatures":0,"replication_records":0,"graph_snapshot":true,"references_resolved":false,"formal_links_present":false},"canonical_record":{"source":{"id":"2306.04406","kind":"arxiv","version":2},"metadata":{"license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","primary_cat":"cs.LG","submitted_at":"2023-06-07T13:04:34Z","cross_cats_sorted":["cs.AI","math.DS","nlin.CD"],"title_canon_sha256":"272048ad51567cb8ea0ab82ad57be66ea2a148434f66e254f45b8bb0c6c7fd64","abstract_canon_sha256":"3ae41b1d5ec177b3e47bd8914f5b6279294bdb9d235803b8f78361e1de38b6b7"},"schema_version":"1.0"},"receipt":{"kind":"pith_receipt","key_id":"pith-v1-2026-05","algorithm":"ed25519","signed_at":"2026-07-05T07:05:44.222025Z","signature_b64":"a1vz9+bpVZCTAE5fn+PCGQGsXMSSYOic5jqwI4DhHstbi5CJ8EpwHWqxDs2dzNd8lNW82fIyH4V57P4rmnpBDQ==","signed_message":"canonical_sha256_bytes","builder_version":"pith-number-builder-2026-05-17-v1","receipt_version":"0.3","canonical_sha256":"88d4e5ce97061c0027178d04c62ab1f38e19423d50f958bbf83e5f493e44f291","last_reissued_at":"2026-07-05T07:05:44.221549Z","signature_status":"signed_v1","first_computed_at":"2026-07-05T07:05:44.221549Z","public_key_fingerprint":"8d4b5ee74e4693bcd1df2446408b0d54"},"graph_snapshot":{"paper":{"title":"Generalized Teacher Forcing for Learning Chaotic Dynamics","license":"http://arxiv.org/licenses/nonexclusive-distrib/1.0/","headline":"","cross_cats":["cs.AI","math.DS","nlin.CD"],"primary_cat":"cs.LG","authors_text":"Daniel Durstewitz, Florian Hess, Manuel Brenner, Zahra Monfared","submitted_at":"2023-06-07T13:04:34Z","abstract_excerpt":"Chaotic dynamical systems (DS) are ubiquitous in nature and society. Often we are interested in reconstructing such systems from observed time series for prediction or mechanistic insight, where by reconstruction we mean learning geometrical and invariant temporal properties of the system in question (like attractors). However, training reconstruction algorithms like recurrent neural networks (RNNs) on such systems by gradient-descent based techniques faces severe challenges. This is mainly due to exploding gradients caused by the exponential divergence of trajectories in chaotic systems. More"},"claims":{"count":0,"items":[],"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"source":{"id":"2306.04406","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/2306.04406/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"},"aliases":[{"alias_kind":"arxiv","alias_value":"2306.04406","created_at":"2026-07-05T07:05:44.221601+00:00"},{"alias_kind":"arxiv_version","alias_value":"2306.04406v2","created_at":"2026-07-05T07:05:44.221601+00:00"},{"alias_kind":"doi","alias_value":"10.48550/arxiv.2306.04406","created_at":"2026-07-05T07:05:44.221601+00:00"},{"alias_kind":"pith_short_12","alias_value":"RDKOLTUXAYOA","created_at":"2026-07-05T07:05:44.221601+00:00"},{"alias_kind":"pith_short_16","alias_value":"RDKOLTUXAYOAAJYX","created_at":"2026-07-05T07:05:44.221601+00:00"},{"alias_kind":"pith_short_8","alias_value":"RDKOLTUX","created_at":"2026-07-05T07:05:44.221601+00:00"}],"events":[],"event_summary":{},"paper_claims":[],"inbound_citations":{"count":4,"internal_anchor_count":0,"sample":[{"citing_arxiv_id":"2605.31438","citing_title":"Flow map learning in nonlinear vector autoregressive models: influence of the feature-library structure on the training error","ref_index":69,"is_internal_anchor":false},{"citing_arxiv_id":"2605.04024","citing_title":"Can Transformers predict system collapse in dynamical systems?","ref_index":56,"is_internal_anchor":false},{"citing_arxiv_id":"2604.25904","citing_title":"Teacher Forcing as Generalized Bayes: Optimization Geometry Mismatch in Switching Surrogates for Chaotic Dynamics","ref_index":3,"is_internal_anchor":false},{"citing_arxiv_id":"2605.01439","citing_title":"Optimizing Reservoir Computing for Reconstructing Ergodic Properties","ref_index":27,"is_internal_anchor":false}]},"formal_canon":{"evidence_count":0,"sample":[],"anchors":[]},"links":{"html":"https://pith.science/pith/RDKOLTUXAYOAAJYXRUCMMKVR6O","json":"https://pith.science/pith/RDKOLTUXAYOAAJYXRUCMMKVR6O.json","graph_json":"https://pith.science/api/pith-number/RDKOLTUXAYOAAJYXRUCMMKVR6O/graph.json","events_json":"https://pith.science/api/pith-number/RDKOLTUXAYOAAJYXRUCMMKVR6O/events.json","paper":"https://pith.science/paper/RDKOLTUX"},"agent_actions":{"view_html":"https://pith.science/pith/RDKOLTUXAYOAAJYXRUCMMKVR6O","download_json":"https://pith.science/pith/RDKOLTUXAYOAAJYXRUCMMKVR6O.json","view_paper":"https://pith.science/paper/RDKOLTUX","resolve_alias":"https://pith.science/api/pith-number/resolve?arxiv=2306.04406&json=true","fetch_graph":"https://pith.science/api/pith-number/RDKOLTUXAYOAAJYXRUCMMKVR6O/graph.json","fetch_events":"https://pith.science/api/pith-number/RDKOLTUXAYOAAJYXRUCMMKVR6O/events.json","actions":{"anchor_timestamp":"https://pith.science/pith/RDKOLTUXAYOAAJYXRUCMMKVR6O/action/timestamp_anchor","attest_storage":"https://pith.science/pith/RDKOLTUXAYOAAJYXRUCMMKVR6O/action/storage_attestation","attest_author":"https://pith.science/pith/RDKOLTUXAYOAAJYXRUCMMKVR6O/action/author_attestation","sign_citation":"https://pith.science/pith/RDKOLTUXAYOAAJYXRUCMMKVR6O/action/citation_signature","submit_replication":"https://pith.science/pith/RDKOLTUXAYOAAJYXRUCMMKVR6O/action/replication_record"}},"created_at":"2026-07-05T07:05:44.221601+00:00","updated_at":"2026-07-05T07:05:44.221601+00:00"}