{"paper":{"title":"Quantifying information flow along a stochastic trajectory","license":"http://creativecommons.org/licenses/by/4.0/","headline":"Deep learning enables estimation of information flow along individual stochastic trajectories from time-series data.","cross_cats":[],"primary_cat":"cond-mat.stat-mech","authors_text":"Euijoon Kwon, Yongjae Oh, Yongjoo Baek","submitted_at":"2026-05-13T13:28:17Z","abstract_excerpt":"Stochastic information flow (SIF) quantifies information flow at the trajectory level, overcoming the limitations of conventional symmetric, ensemble-averaged measures. However, computational difficulties have hindered the empirical application of the SIF. In this work, we propose a scalable deep-learning method for estimating the SIF from general time-series data. Its applications to an exactly solvable two-particle model, Kuramoto oscillators, and empirical trajectories of interacting motile cells demonstrate the utility of SIF as a data-driven indicator of cooperative structures."},"claims":{"count":4,"items":[{"kind":"strongest_claim","text":"we propose a scalable deep-learning method for estimating the SIF from general time-series data","source":"verdict.strongest_claim","status":"machine_extracted","claim_id":"C1","attestation":"unclaimed"},{"kind":"weakest_assumption","text":"That a neural network trained on simulated or limited data can accurately recover the true trajectory-level information flow for arbitrary unseen stochastic processes without model-specific tuning or overfitting.","source":"verdict.weakest_assumption","status":"machine_extracted","claim_id":"C2","attestation":"unclaimed"},{"kind":"one_line_summary","text":"A scalable deep-learning estimator for trajectory-level stochastic information flow is proposed and tested on solvable models, oscillators, and motile cell trajectories.","source":"verdict.one_line_summary","status":"machine_extracted","claim_id":"C3","attestation":"unclaimed"},{"kind":"headline","text":"Deep learning enables estimation of information flow along individual stochastic trajectories from time-series data.","source":"verdict.pith_extraction.headline","status":"machine_extracted","claim_id":"C4","attestation":"unclaimed"}],"snapshot_sha256":"7858dbb0ad3abb83a5a5a3e3f816e88648b451d73777ac160bae01383a871294"},"source":{"id":"2605.13509","kind":"arxiv","version":1},"verdict":{"id":"375767c3-ebf2-4c7c-85dd-aac30a570e0a","model_set":{"reader":"grok-4.3"},"created_at":"2026-05-14T19:14:44.078437Z","strongest_claim":"we propose a scalable deep-learning method for estimating the SIF from general time-series data","one_line_summary":"A scalable deep-learning estimator for trajectory-level stochastic information flow is proposed and tested on solvable models, oscillators, and motile cell trajectories.","pipeline_version":"pith-pipeline@v0.9.0","weakest_assumption":"That a neural network trained on simulated or limited data can accurately recover the true trajectory-level information flow for arbitrary unseen stochastic processes without model-specific tuning or overfitting.","pith_extraction_headline":"Deep learning enables estimation of information flow along individual stochastic trajectories from time-series data."},"references":{"count":75,"sample":[{"doi":"","year":1998,"title":"Sekimoto, Langevin equation and thermodynamics, Prog","work_id":"f2c14190-f09b-492d-92f4-fb9fd5023ed8","ref_index":1,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":2012,"title":"Seifert, Stochastic thermodynamics, fluctuation the- orems and molecular machines, Rep","work_id":"e931af35-ce5f-4e92-b781-18fe887de15c","ref_index":2,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1993,"title":"D. J. Evans, E. G. D. Cohen, and G. P. Morriss, Proba- bility of second law violations in shearing steady states, Phys. Rev. Lett.71, 2401 (1993)","work_id":"84bd0944-fc78-4ad4-982a-7b8efd9373b8","ref_index":3,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1997,"title":"Jarzynski, Nonequilibrium equality for free energy dif- ferences, Phys","work_id":"38fff1ff-d736-4d71-ad8d-a614c3d205b9","ref_index":4,"cited_arxiv_id":"","is_internal_anchor":false},{"doi":"","year":1999,"title":"G. E. Crooks, Entropy production fluctuation theorem and the nonequilibrium work relation for free energy dif- ferences, Phys. Rev. E60, 2721 (1999)","work_id":"37734ff5-1b1d-4d39-9adb-f37d60467ea8","ref_index":5,"cited_arxiv_id":"","is_internal_anchor":false}],"resolved_work":75,"snapshot_sha256":"b27f5d6dd1501793acee40780d9257dd07f68c87cbbfcbbcec20f43bac069bbd","internal_anchors":0},"formal_canon":{"evidence_count":1,"snapshot_sha256":"141e76e57132f1a9a7ceae25cf37627ae83f0fe91ed43b5b0586e7525900e55c"},"author_claims":{"count":0,"strong_count":0,"snapshot_sha256":"258153158e38e3291e3d48162225fcdb2d5a3ed65a07baac614ab91432fd4f57"},"builder_version":"pith-number-builder-2026-05-17-v1"}