{"status":"ok","message-type":"work","message-version":"1.0.0","message":{"indexed":{"date-parts":[[2025,12,19]],"date-time":"2025-12-19T10:10:51Z","timestamp":1766139051811,"version":"build-2065373602"},"reference-count":83,"publisher":"Institute of Electrical and Electronics Engineers (IEEE)","issue":"3","license":[{"start":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T00:00:00Z","timestamp":1740787200000},"content-version":"vor","delay-in-days":0,"URL":"https:\/\/ieeexplore.ieee.org\/Xplorehelp\/downloads\/license-information\/IEEE.html"},{"start":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T00:00:00Z","timestamp":1740787200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-029"},{"start":{"date-parts":[[2025,3,1]],"date-time":"2025-03-01T00:00:00Z","timestamp":1740787200000},"content-version":"stm-asf","delay-in-days":0,"URL":"https:\/\/doi.org\/10.15223\/policy-037"}],"funder":[{"DOI":"10.13039\/501100015401","name":"Key Research and Development Project in Shaanxi Province","doi-asserted-by":"publisher","award":["2022GXLH-01-03"],"award-info":[{"award-number":["2022GXLH-01-03"]}],"id":[{"id":"10.13039\/501100015401","id-type":"DOI","asserted-by":"publisher"}]},{"DOI":"10.13039\/501100001809","name":"National Natural Science Foundation of China","doi-asserted-by":"publisher","award":["62172326"],"award-info":[{"award-number":["62172326"]}],"id":[{"id":"10.13039\/501100001809","id-type":"DOI","asserted-by":"publisher"}]}],"content-domain":{"domain":[],"crossmark-restriction":false},"short-container-title":["IEEE Trans. Inform. Theory"],"published-print":{"date-parts":[[2025,3]]},"DOI":"10.1109\/tit.2025.3531136","type":"journal-article","created":{"date-parts":[[2025,1,17]],"date-time":"2025-01-17T18:34:48Z","timestamp":1737138888000},"page":"2028-2053","source":"Crossref","is-referenced-by-count":3,"title":["How Does Distribution Matching Help Domain Generalization: An Information-Theoretic Analysis"],"prefix":"10.1109","volume":"71","author":[{"ORCID":"https:\/\/orcid.org\/0000-0002-4475-5056","authenticated-orcid":false,"given":"Yuxin","family":"Dong","sequence":"first","affiliation":[{"name":"School of Computer Science and Technology, Xi’an Jiaotong University, Xi’an, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-3840-441X","authenticated-orcid":false,"given":"Tieliang","family":"Gong","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Xi’an Jiaotong University, Xi’an, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Hong","family":"Chen","sequence":"additional","affiliation":[{"name":"College of Science, Huazhong Agriculture University, Wuhan, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"given":"Shuangyong","family":"Song","sequence":"additional","affiliation":[{"name":"China Telecom Corporation, Beijing, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0003-0330-5435","authenticated-orcid":false,"given":"Weizhan","family":"Zhang","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Xi’an Jiaotong University, Xi’an, China"}],"role":[{"role":"author","vocabulary":"crossref"}]},{"ORCID":"https:\/\/orcid.org\/0000-0002-0079-3106","authenticated-orcid":false,"given":"Chen","family":"Li","sequence":"additional","affiliation":[{"name":"School of Computer Science and Technology, Xi’an Jiaotong University, Xi’an, China"}],"role":[{"role":"author","vocabulary":"crossref"}]}],"member":"263","reference":[{"article-title":"ImageNet-trained CNNs are biased towards texture; increasing shape bias improves accuracy and robustness","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Geirhos","key":"ref1"},{"article-title":"Benchmarking neural network robustness to common corruptions and perturbations","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Hendrycks","key":"ref2"},{"key":"ref3","first-page":"1","article-title":"Why do deep convolutional networks generalize so poorly to small image transformations","volume":"20","author":"Azulay","year":"2019","journal-title":"J. Mach. Learn. Res."},{"key":"ref4","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00823"},{"key":"ref5","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-319-49409-8_35"},{"key":"ref6","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2018.00566"},{"key":"ref7","doi-asserted-by":"publisher","DOI":"10.5555\/2946645.2946704"},{"key":"ref8","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01267-0_38"},{"key":"ref9","article-title":"Invariant risk minimization","author":"Arjovsky","year":"2019","journal-title":"arXiv:1907.02893"},{"key":"ref10","article-title":"Invariant causal mechanisms through distribution matching","author":"Chevalley","year":"2022","journal-title":"arXiv:2206.11646"},{"journal-title":"arXiv:2008.01883","article-title":"When is invariance useful in an out-of-distribution generalization problem?","author":"Koyama","key":"ref11"},{"article-title":"Gradient matching for domain generalization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Shi","key":"ref12"},{"key":"ref13","first-page":"18347","article-title":"Fishr: Invariant gradient variances for out-of-distribution generalization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Rame"},{"key":"ref14","first-page":"1","article-title":"Distributionally robust neural networks","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Sagawa"},{"key":"ref15","first-page":"5815","article-title":"Out-of-distribution generalization via risk extrapolation (REx)","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Krueger"},{"key":"ref16","first-page":"17340","article-title":"Probable domain generalization via quantile risk minimization","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Eastwood"},{"issue":"1","key":"ref17","first-page":"46","article-title":"Domain generalization by marginal transfer learning","volume":"22","author":"Blanchard","year":"2017","journal-title":"J. Mach. Learn. Res."},{"key":"ref18","first-page":"23664","article-title":"Adaptive risk minimization: Learning to adapt to domain shift","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Zhang"},{"article-title":"Understanding the failure modes of out-of-distribution generalization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Nagarajan","key":"ref19"},{"article-title":"Learning explanations that are hard to vary","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Parascandolo","key":"ref20"},{"article-title":"In search of lost domain generalization","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Gulrajani","key":"ref21"},{"key":"ref22","first-page":"2524","article-title":"Information-theoretic analysis of generalization capability of learning algorithms","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"30","author":"Xu"},{"key":"ref23","first-page":"11013","article-title":"Information-theoretic generalization bounds for SGLD via data-dependent estimates","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"32","author":"Negrea"},{"key":"ref24","first-page":"3526","article-title":"Information-theoretic generalization bounds for stochastic gradient descent","volume-title":"Proc. Conf. Learn. Theory","author":"Neu"},{"article-title":"On the generalization of models trained with SGD: Information-theoretic bounds and implications","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Wang","key":"ref25"},{"key":"ref26","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2024.3365728"},{"key":"ref27","first-page":"1225","article-title":"Train faster, generalize better: Stability of stochastic gradient descent","volume-title":"Proc. 33rd Int. Conf. Mach. Learn.","author":"Hardt"},{"key":"ref28","first-page":"4381","article-title":"Stability of stochastic gradient descent on nonsmooth convex losses","volume-title":"Proc. Int. Conf. Adv. Neural Inf. Process. Syst.","volume":"33","author":"Bassily"},{"key":"ref29","first-page":"6175","article-title":"Stability and generalization of stochastic gradient methods for minimax problems","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Lei"},{"key":"ref30","first-page":"19109","article-title":"Tighter expected generalization error bounds via Wasserstein distance","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","author":"Rodr\u00edguez-G\u00e1lvez"},{"key":"ref31","first-page":"20160","article-title":"Simple stochastic and online gradient descent algorithms for pairwise learning","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Yang"},{"key":"ref32","first-page":"2026","article-title":"Stability and differential privacy of stochastic gradient descent for pairwise learning with non-smooth loss","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Yang"},{"article-title":"Domain adaptation: Learning bounds and algorithms","volume-title":"Proc. 22nd Annu. Conf. Learn. Theory (COLT)","author":"Mansour","key":"ref33"},{"key":"ref34","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11784"},{"article-title":"Information-theoretic analysis of unsupervised domain adaptation","volume-title":"Proc. The 11th Int. Conf. Learn. Represent.","author":"Wang","key":"ref35"},{"key":"ref36","first-page":"3438","article-title":"Invariance principle meets information bottleneck for out-of-distribution generalization","volume-title":"Proc. Annu. Conf. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Ahuja"},{"key":"ref37","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01555"},{"key":"ref38","first-page":"27222","article-title":"Sparse invariant risk minimization","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Zhou"},{"key":"ref39","doi-asserted-by":"publisher","DOI":"10.1109\/TPAMI.2021.3094760"},{"key":"ref40","first-page":"10108","article-title":"A new family of generalization bounds using samplewise evaluated CMI","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Hellstr\u00f6m"},{"key":"ref41","first-page":"36111","article-title":"Tighter information-theoretic generalization bounds from supersamples","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Wang"},{"key":"ref42","first-page":"17628","article-title":"An information-theoretic approach to distribution shifts","volume-title":"Proc. NIPS","volume":"34","author":"Federici"},{"key":"ref43","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00116"},{"key":"ref44","doi-asserted-by":"publisher","DOI":"10.1109\/JSAIT.2020.2991139"},{"key":"ref45","first-page":"24670","article-title":"Information-theoretic generalization bounds for black-box learning algorithms","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Harutyunyan"},{"article-title":"KL guided domain adaptation","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Nguyen","key":"ref46"},{"key":"ref47","article-title":"SAND-Mask: An enhanced gradient masking strategy for the discovery of invariances in domain generalization","author":"Shahtalebi","year":"2021","journal-title":"arXiv:2106.02266"},{"key":"ref48","first-page":"7523","article-title":"On learning invariant representations for domain adaptation","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Zhao"},{"key":"ref49","first-page":"4069","article-title":"Does invariant risk minimization capture invariance","volume-title":"Proc. Int. Conf. Artif. Intell. Statist.","author":"Kamath"},{"key":"ref50","doi-asserted-by":"publisher","DOI":"10.1109\/TIT.2019.2945779"},{"key":"ref51","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT50566.2022.9834474"},{"key":"ref52","first-page":"8106","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":"ref53","doi-asserted-by":"publisher","DOI":"10.3390\/e22090999"},{"article-title":"Learning robust representations via multi-view information bottleneck","volume-title":"Proc. Int. Conf. Learn. Represent.","author":"Federici","key":"ref54"},{"key":"ref55","first-page":"19538","article-title":"Compressive visual representations","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Lee"},{"key":"ref56","first-page":"1","article-title":"Rethinking information-theoretic generalization: Loss entropy induced PAC bounds","volume-title":"Proc. The 12th Int. Conf. Learn. Represent.","author":"Dong"},{"key":"ref57","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT.2018.8437571"},{"key":"ref58","first-page":"24222","article-title":"Analyzing the generalization capability of SGLD using properties of Gaussian channels","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Wang"},{"key":"ref59","first-page":"25878","article-title":"Generalization bounds for meta-learning: An information-theoretic analysis","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Chen"},{"key":"ref60","doi-asserted-by":"publisher","DOI":"10.3390\/e23010126"},{"key":"ref61","first-page":"20648","article-title":"Evaluated CMI bounds for meta learning: Tightness and expressiveness","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"35","author":"Hellstr\u00f6m"},{"key":"ref62","doi-asserted-by":"publisher","DOI":"10.1109\/ISIT54713.2023.10206566"},{"key":"ref63","doi-asserted-by":"publisher","DOI":"10.1561\/2200000112"},{"key":"ref64","doi-asserted-by":"publisher","DOI":"10.1017\/9781108966351"},{"key":"ref65","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2015.293"},{"key":"ref66","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2013.208"},{"key":"ref67","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2017.591"},{"key":"ref68","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR.2017.572"},{"key":"ref69","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-01270-0_28"},{"key":"ref70","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV.2019.00149"},{"key":"ref71","first-page":"22405","article-title":"SWAD: Domain generalization by seeking flat minima","volume-title":"Proc. Adv. Neural Inf. Process. Syst.","volume":"34","author":"Cha"},{"article-title":"Causal balancing for domain generalization","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Wang","key":"ref72"},{"article-title":"PGrad: Learning principal gradients for domain generalization","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Wang","key":"ref73"},{"article-title":"Bitrate-constrained DRO: Beyond worst case robustness to unknown group shifts","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Setlur","key":"ref74"},{"article-title":"Pareto invariant risk minimization: Towards mitigating the optimization dilemma in out-of-distribution generalization","volume-title":"Proc. 11th Int. Conf. Learn. Represent.","author":"Chen","key":"ref75"},{"key":"ref76","article-title":"Improve unsupervised domain adaptation with mixup training","author":"Yan","year":"2020","journal-title":"arXiv:2001.00677"},{"key":"ref77","doi-asserted-by":"publisher","DOI":"10.1609\/aaai.v32i1.11596"},{"key":"ref78","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR46437.2021.00858"},{"key":"ref79","doi-asserted-by":"publisher","DOI":"10.1007\/978-3-030-58536-5_8"},{"key":"ref80","doi-asserted-by":"publisher","DOI":"10.1109\/ICCV48922.2021.00948"},{"key":"ref81","first-page":"17848","article-title":"Adaptive second order coresets for data-efficient machine learning","volume-title":"Proc. Int. Conf. Mach. Learn.","author":"Pooladzandi"},{"issue":"1","key":"ref82","first-page":"10237","article-title":"Underspecification presents challenges for credibility in modern machine learning","volume":"23","author":"D\u2019Amour","year":"2020","journal-title":"J. Mach. Learn. Res."},{"key":"ref83","doi-asserted-by":"publisher","DOI":"10.1109\/CVPR52688.2022.01626"}],"container-title":["IEEE Transactions on Information Theory"],"original-title":[],"link":[{"URL":"http:\/\/xplorestaging.ieee.org\/ielx8\/18\/10896902\/10844895.pdf?arnumber=10844895","content-type":"unspecified","content-version":"vor","intended-application":"similarity-checking"}],"deposited":{"date-parts":[[2025,2,21]],"date-time":"2025-02-21T12:12:19Z","timestamp":1740139939000},"score":1,"resource":{"primary":{"URL":"https:\/\/ieeexplore.ieee.org\/document\/10844895\/"}},"subtitle":[],"short-title":[],"issued":{"date-parts":[[2025,3]]},"references-count":83,"journal-issue":{"issue":"3"},"URL":"https:\/\/doi.org\/10.1109\/tit.2025.3531136","relation":{},"ISSN":["0018-9448","1557-9654"],"issn-type":[{"type":"print","value":"0018-9448"},{"type":"electronic","value":"1557-9654"}],"subject":[],"published":{"date-parts":[[2025,3]]}}}