{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2026,3,20]],"date-time":"2026-03-20T03:08:08Z","timestamp":1773976088736,"version":"3.50.1"},"reference-count":93,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"1","license":[{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"am","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2024,1,1]],"date-time":"2024-01-01T00:00:00Z","timestamp":1704067200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100000266","name":"Engineering and Physical Sciences Research Council","doi-asserted-by":"publisher","award":["EP\/V056883\/1"],"award-info":[{"award-number":["EP\/V056883\/1"]}],"id":[{"id":"10.13039\/501100000266","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100000001","name":"National Science Foundation","doi-asserted-by":"publisher","award":["CCF-1816209"],"award-info":[{"award-number":["CCF-1816209"]}],"id":[{"id":"10.13039\/100000001","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100000288","name":"Royal Society","doi-asserted-by":"publisher","award":["NIFR1 192656"],"award-info":[{"award-number":["NIFR1 192656"]}],"id":[{"id":"10.13039\/501100000288","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/100006602","name":"Air Force Research Laboratory","doi-asserted-by":"publisher","award":["FA8750-19-2-1000"],"award-info":[{"award-number":["FA8750-19-2-1000"]}],"id":[{"id":"10.13039\/100006602","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Inform. Theory"],"published-print":{"date-parts":[[2024,1]]},"DOI":"10.1109\/tit.2023.3329617","type":"journal-article","created":{"date-parts":[[2023,11,2]],"date-time":"2023-11-02T17:49:02Z","timestamp":1698947342000},"page":"632-655","source":"Crossref","is-referenced-by-count":9,"title":["Information-Theoretic Characterizations of Generalization Error for the Gibbs Algorithm"],"prefix":"10.1109","volume":"70","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4761-0151","authenticated-orcid":false,"given":"Gholamali","family":"Aminian","sequence":"first","affiliation":[{"name":"The Alan Turing Institute, London, U.K."}]},{"given":"Yuheng","family":"Bu","sequence":"additional","affiliation":[{"name":"Department of Electrical Engineering and Computer Science, Massachusetts Institute of Technology, Cambridge, MA, USA"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8441-8791","authenticated-orcid":false,"given":"Laura","family":"Toni","sequence":"additional","affiliation":[{"name":"Electronic and Electrical Engineering Department, University College London, London, U.K."}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-8908-847X","authenticated-orcid":false,"given":"Miguel R. D.","family":"Rodrigues","sequence":"additional","affiliation":[{"name":"Electronic and Electrical Engineering Department, University College London, London, U.K."}]},{"ORCID":"https:\/\/orcid.org\/0000-0001-9166-4758","authenticated-orcid":false,"given":"Gregory W.","family":"Wornell","sequence":"additional","affiliation":[{"name":"The Alan Turing Institute, London, U.K."}]}],"member":"263","reference":[{"key":"ref1","article-title":"An exact characterization of the generalization error for the Gibbs algorithm","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Aminian"},{"key":"ref2","doi-asserted-by":"publisher","DOI":"10.1017\/9781108616799"},{"key":"ref3","doi-asserted-by":"publisher","DOI":"10.1109\/72.788640"},{"key":"ref4","first-page":"499","article-title":"Stability and generalization","volume":"2","author":"Bousquet","year":"2002","journal-title":"J. Mach. Learn. Res."},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-011-5268-1"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1023\/A:1021840411064"},{"key":"ref7","first-page":"2524","article-title":"Information-theoretic analysis of generalization capability of learning algorithms","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Xu"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1017\/cbo9780511624216"},{"key":"ref9","doi-asserted-by":"publisher","DOI":"10.1145\/3446776"},{"key":"ref10","doi-asserted-by":"publisher","DOI":"10.1098\/rspa.1946.0056"},{"key":"ref11","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2007.915715"},{"key":"ref12","doi-asserted-by":"publisher","DOI":"10.1109\/TMBMC.2015.2502858"},{"key":"ref13","article-title":"Pac-Bayesian supervised classification: The thermodynamics of statistical learning","author":"Catoni","year":"2007","journal-title":"arXiv:0712.0248"},{"key":"ref14","doi-asserted-by":"publisher","DOI":"10.1038\/066291a0"},{"key":"ref15","doi-asserted-by":"publisher","DOI":"10.1103\/PhysRev.106.620"},{"key":"ref16","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2005.864439"},{"key":"ref17","article-title":"Empirical risk minimization with generalized relative entropy regularization","author":"Perlaza","year":"2022","journal-title":"arXiv:2211.06617"},{"key":"ref18","doi-asserted-by":"publisher","DOI":"10.1214\/009053606000000704"},{"key":"ref19","first-page":"2028","article-title":"Distributiondependent analysis of Gibbs-ERM principle","volume-title":"Proc. Conf. Learn. Theory","author":"Kuzborskij"},{"key":"ref20","doi-asserted-by":"publisher","DOI":"10.1109\/itw.2016.7606789"},{"key":"ref21","first-page":"466","article-title":"A strongly quasiconvex PAC-Bayesian bound","volume-title":"Proc. Int. Conf. Algorithmic Learn. Theory","author":"Thiemann"},{"key":"ref22","article-title":"PAC-Bayes mini-tutorial: A continuous union bound","author":"van Erven","year":"2014","journal-title":"arXiv:1405.1580"},{"key":"ref23","doi-asserted-by":"publisher","DOI":"10.1137\/0325042"},{"key":"ref24","first-page":"1","article-title":"On the trend to equilibrium for the Fokker-Planck equation: An interplay between physics and functional analysis","volume":"19","author":"Markowich","year":"2000","journal-title":"Mat. Contemp"},{"key":"ref25","first-page":"1674","article-title":"Non-convex learning via stochastic gradient Langevin dynamics: A nonasymptotic analysis","volume-title":"Proc. Conf. Learn. Theory","author":"Raginsky"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2019.2945779"},{"key":"ref27","doi-asserted-by":"publisher","DOI":"10.1109\/JSAIT.2020.2991139"},{"key":"ref28","first-page":"7234","article-title":"Chaining mutual information and tightening generalization bounds","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Asadi"},{"issue":"139","key":"ref29","first-page":"1","article-title":"Chaining meets chain rule: Multilevel entropic regularization and training of neural networks","volume":"21","author":"Asadi","year":"2020","journal-title":"J. Mach. Learn. Res."},{"key":"ref30","first-page":"3437","article-title":"Reasoning about generalization via conditional mutual information","volume-title":"Proc. Conf. Learn. Theory","author":"Steinke"},{"key":"ref31","article-title":"Conditioning and processing: Techniques to improve information-theoretic generalization bounds","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Hafez-Kolahi"},{"key":"ref32","article-title":"Sharpened generalization bounds based on conditional mutual information and an application to noisy, iterative algorithms","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Haghifam"},{"key":"ref33","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT.2017.8006774"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1109\/isit.2019.8849834"},{"key":"ref35","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2021.3085190"},{"key":"ref36","doi-asserted-by":"publisher","DOI":"10.1109\/ITW46852.2021.9457642"},{"key":"ref37","article-title":"Learning algorithm generalization error bounds via auxiliary distributions","author":"Aminian","year":"2022","journal-title":"arXiv:2210.00483"},{"key":"ref38","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT50566.2022.9834474"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/ITW.2018.8613445"},{"key":"ref40","doi-asserted-by":"publisher","DOI":"10.1109\/isit.2019.8849359"},{"key":"ref41","first-page":"19109","article-title":"Tighter expected generalization error bounds via Wasserstein distance","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"G\u00e1lvez"},{"key":"ref42","article-title":"A unified framework for information-theoretic generalization bounds","author":"Chu","year":"2023","journal-title":"arXiv:2305.11042"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT45174.2021.9517732"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v34i04.5730"},{"key":"ref45","doi-asserted-by":"publisher","DOI":"10.3390\/e23101255"},{"key":"ref46","first-page":"7433","article-title":"An information-theoretical approach to semi-supervised learning under covariate-shift","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Aminian"},{"key":"ref47","first-page":"8494","article-title":"How does pseudo-labeling affect the generalization error of the semi-supervised Gibbs algorithm?","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"He"},{"key":"ref48","first-page":"8673","article-title":"Characterizing and understanding the generalization error of transfer learning with Gibbs algorithm","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Bu"},{"key":"ref49","doi-asserted-by":"publisher","DOI":"10.3390\/e23010126"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT54713.2023.10206566"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1145\/267460.267466"},{"key":"ref52","doi-asserted-by":"publisher","DOI":"10.1145\/279943.279989"},{"key":"ref53","article-title":"User-friendly introduction to PAC-Bayes bounds","author":"Alquier","year":"2021","journal-title":"arXiv:2110.11216"},{"key":"ref54","first-page":"1711","article-title":"Novel change of measure inequalities with applications to PAC-Bayesian bounds and Monte Carlo estimation","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Ohnishi"},{"key":"ref55","doi-asserted-by":"publisher","DOI":"10.1007\/s10994-017-5690-0"},{"key":"ref56","first-page":"435","article-title":"PAC-Bayesian bounds based on the R\u00e9nyi divergence","volume-title":"Proc. Artif. Intell. Statist.","author":"B\u00e9gin"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/JSAIT.2020.3040992"},{"key":"ref58","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT45174.2021.9518043"},{"key":"ref59","article-title":"PACBayes analysis beyond the usual bounds","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Rivasplata"},{"key":"ref60","first-page":"8440","article-title":"Data-dependent PAC-Bayes priors via differential privacy","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Dziugaite"},{"key":"ref61","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT45174.2021.9517960"},{"key":"ref62","first-page":"1377","article-title":"Entropy-SGD optimizes the prior of a PAC-Bayes bound: Generalization properties of entropy-SGD and data-dependent priors","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Dziugaite"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.7551\/mitpress\/7503.003.0007"},{"key":"ref64","article-title":"A PAC-Bayesian approach to adaptive classification","volume":"840","author":"Catoni","year":"2003"},{"issue":"1","key":"ref65","first-page":"3507","article-title":"PAC-Bayes bounds with data dependent priors","volume":"13","author":"Parrado-Hern\u00e1ndez","year":"2012","journal-title":"J. Mach. Learn. Res."},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1016\/j.tcs.2012.10.013"},{"key":"ref67","article-title":"PAC-Bayes bounds for stable algorithms with instancedependent priors","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Rivasplata"},{"issue":"4","key":"ref68","first-page":"787","article-title":"Risk bounds for the majority vote: From a PAC-Bayesian analysis to a learning algorithm","volume":"16","author":"Germain","year":"2015","journal-title":"J. Mach. Learn. Res."},{"key":"ref69","first-page":"1884","article-title":"PAC-Bayesian theory meets Bayesian inference","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Germain"},{"issue":"1","key":"ref70","first-page":"8374","article-title":"On the properties of variational approximations of Gibbs posteriors","volume":"17","author":"Alquier","year":"2016","journal-title":"J. Mach. Learn. Res."},{"key":"ref71","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT54713.2023.10206506"},{"key":"ref72","article-title":"Mean-field analysis of generalization errors","author":"Aminian","year":"2023","journal-title":"arXiv:2306.11623"},{"key":"ref73","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2016.2603151"},{"issue":"2","key":"ref74","first-page":"16","article-title":"Conjugate Bayesian analysis of the Gaussian distribution","volume":"1","author":"Murphy","year":"2007","journal-title":"Def"},{"key":"ref75","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-45381-1_10"},{"key":"ref76","doi-asserted-by":"publisher","DOI":"10.1109\/FOCS.2007.66"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1214\/aop\/1176994579"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1007\/s13171-010-0006-5"},{"key":"ref79","article-title":"Wasserstein convergence rates of increasingly concentrating probability measures","author":"Hasenpflug","year":"2022","journal-title":"arXiv:2207.08551"},{"key":"ref80","volume-title":"Information Theory: From Coding to Learning","author":"Polyanskiy","year":"2022"},{"key":"ref81","doi-asserted-by":"publisher","DOI":"10.1017\/cbo9780511802256"},{"key":"ref82","doi-asserted-by":"publisher","DOI":"10.1007\/978-1-4612-1694-0_15"},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1214\/12-EJS675"},{"key":"ref84","first-page":"14973","article-title":"Adversarial training is a form of data-dependent operator norm regularization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Roth"},{"key":"ref85","doi-asserted-by":"publisher","DOI":"10.11650\/twjm\/1500406018"},{"key":"ref86","first-page":"351","article-title":"Dropout training as adaptive regularization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Wager"},{"issue":"2","key":"ref87","first-page":"1","article-title":"Maximum relative margin and datadependent regularization","volume":"11","author":"Shivaswamy","year":"2010","journal-title":"J. Mach. Learn. Res."},{"key":"ref88","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2014.2320500"},{"key":"ref89","doi-asserted-by":"publisher","DOI":"10.1093\/acprof:oso\/9780199535255.001.0001"},{"key":"ref90","article-title":"Concentration inequalities for statistical inference","author":"Zhang","year":"2020","journal-title":"arXiv:2011.02258"},{"key":"ref91","first-page":"31","article-title":"An exponential Efron-Stein inequality for Lq stable learning rules","volume-title":"Proc. Algorithmic Learn. Theory","author":"Abou-Moustafa"},{"key":"ref92","article-title":"Still no free lunches: The price to pay for tighter PAC-Bayes bounds","author":"Guedj","year":"2019","journal-title":"arXiv:1910.04460"},{"key":"ref93","doi-asserted-by":"publisher","DOI":"10.1017\/9781108231596"}],"container-title":["IEEE Transactions on Information Theory"],"original-title":[],"link":[{"URL":"https:\/\/ieeexplore.ieee.org\/ielam\/18\/10375320\/10305226-aam.pdf","content-type":"application\/pdf","content-version":"am","intended-application":"syndication"},{"URL":"http:\/\/xplorestaging.ieee.org\/ielx7\/18\/10375320\/10305226.pdf?arnumber=10305226","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2024,1,12]],"date-time":"2024-01-12T03:48:09Z","timestamp":1705031289000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10305226\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2024,1]]},"references-count":93,"journal-issue":{"issue":"1"},"URL":"https:\/\/doi.org\/10.1109\/tit.2023.3329617","relation":{},"ISSN":["0018-9448","1557-9654"],"issn-type":[{"value":"0018-9448","type":"print"},{"value":"1557-9654","type":"electronic"}],"subject":[],"published":{"date-parts":[[2024,1]]}}}