Information geometryPub Date : 2022-01-01Epub Date: 2021-07-30DOI: 10.1007/s41884-021-00053-7
Ting-Kam Leonard Wong, Jiaowen Yang
{"title":"Pseudo-Riemannian geometry encodes information geometry in optimal transport.","authors":"Ting-Kam Leonard Wong, Jiaowen Yang","doi":"10.1007/s41884-021-00053-7","DOIUrl":"https://doi.org/10.1007/s41884-021-00053-7","url":null,"abstract":"<p><p>Optimal transport and information geometry both study geometric structures on spaces of probability distributions. Optimal transport characterizes the cost-minimizing movement from one distribution to another, while information geometry originates from coordinate invariant properties of statistical inference. Their relations and applications in statistics and machine learning have started to gain more attention. In this paper we give a new differential-geometric relation between the two fields. Namely, the pseudo-Riemannian framework of Kim and McCann, which provides a geometric perspective on the fundamental Ma-Trudinger-Wang (MTW) condition in the regularity theory of optimal transport maps, encodes the dualistic structure of statistical manifold. This general relation is described using the framework of <i>c</i>-divergence under which divergences are defined by optimal transport maps. As a by-product, we obtain a new information-geometric interpretation of the MTW tensor on the graph of the transport map. This relation sheds light on old and new aspects of information geometry. The dually flat geometry of Bregman divergence corresponds to the quadratic cost and the pseudo-Euclidean space, and the logarithmic <math><msup><mi>L</mi> <mrow><mo>(</mo> <mi>α</mi> <mo>)</mo></mrow> </msup> </math> -divergence introduced by Pal and the first author has constant sectional curvature in a sense to be made precise. In these cases we give a geometric interpretation of the information-geometric curvature in terms of the divergence between a primal-dual pair of geodesics.</p>","PeriodicalId":93762,"journal":{"name":"Information geometry","volume":"5 1","pages":"131-159"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9296067/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40535011","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
Information geometryPub Date : 2022-01-01Epub Date: 2022-10-19DOI: 10.1007/s41884-022-00071-z
Eric Smith
{"title":"The information geometry of two-field functional integrals.","authors":"Eric Smith","doi":"10.1007/s41884-022-00071-z","DOIUrl":"https://doi.org/10.1007/s41884-022-00071-z","url":null,"abstract":"<p><p>Two-field functional integrals (2FFI) are an important class of solution methods for generating functions of dissipative processes, including discrete-state stochastic processes, dissipative dynamical systems, and decohering quantum densities. The stationary trajectories of these integrals describe a conserved current by Liouville's theorem, despite the absence of a conserved kinematic phase space current in the underlying stochastic process. We develop the information geometry of generating functions for discrete-state classical stochastic processes in the Doi-Peliti 2FFI form, and exhibit two quantities conserved along stationary trajectories. One is a Wigner function, familiar as a semiclassical density from quantum-mechanical time-dependent density-matrix methods. The second is an overlap function, between directions of variation in an underlying distribution and those in the directions of relative large-deviation probability that can be used to interrogate the distribution, and expressed as an inner product of vector fields in the Fisher information metric. To give an interpretation to the time invertibility implied by current conservation, we use generating functions to represent importance sampling protocols, and show that the conserved Fisher information is the differential of a sample volume under deformations of the nominal distribution and the likelihood ratio. We derive a pair of dual affine connections particular to Doi-Peliti theory for the way they separate the roles of the nominal distribution and likelihood ratio, distinguishing them from the standard dually-flat connection of Nagaoka and Amari defined on the importance distribution, and show that dual flatness in the affine coordinates of the coherent-state basis captures the special role played by coherent states in Doi-Peliti theory.</p>","PeriodicalId":93762,"journal":{"name":"Information geometry","volume":"5 2","pages":"427-492"},"PeriodicalIF":0.0,"publicationDate":"2022-01-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"https://www.ncbi.nlm.nih.gov/pmc/articles/PMC9700636/pdf/","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"40503370","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"OA","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Plücker coordinates of the best-fit Stiefel tropical linear space to a mixture of Gaussian distributions","authors":"K. Miura, R. Yoshida","doi":"10.1007/s41884-023-00098-w","DOIUrl":"https://doi.org/10.1007/s41884-023-00098-w","url":null,"abstract":"","PeriodicalId":93762,"journal":{"name":"Information geometry","volume":"6 1","pages":"171 - 201"},"PeriodicalIF":0.0,"publicationDate":"2021-12-22","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"48081212","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Transport information geometry: Riemannian calculus on probability simplex","authors":"Wuchen Li","doi":"10.1007/s41884-021-00059-1","DOIUrl":"https://doi.org/10.1007/s41884-021-00059-1","url":null,"abstract":"","PeriodicalId":93762,"journal":{"name":"Information geometry","volume":"5 1","pages":"161 - 207"},"PeriodicalIF":0.0,"publicationDate":"2021-11-08","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"43777725","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"The Banach manifold of measures and the Lagrange multipliers of statistical mechanics","authors":"S. Dostoglou, A. Hughes","doi":"10.1007/s41884-021-00057-3","DOIUrl":"https://doi.org/10.1007/s41884-021-00057-3","url":null,"abstract":"","PeriodicalId":93762,"journal":{"name":"Information geometry","volume":"4 1","pages":"377 - 391"},"PeriodicalIF":0.0,"publicationDate":"2021-11-03","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"45397538","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}
{"title":"Riemannian barycentres of Gibbs distributions: new results on concentration and convexity in compact symmetric spaces","authors":"S. Said, J. Manton","doi":"10.1007/s41884-021-00055-5","DOIUrl":"https://doi.org/10.1007/s41884-021-00055-5","url":null,"abstract":"","PeriodicalId":93762,"journal":{"name":"Information geometry","volume":"4 1","pages":"329 - 362"},"PeriodicalIF":0.0,"publicationDate":"2021-09-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"42990385","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}