IEEE Transactions on Pattern Analysis and Machine Intelligence · 2019 · 76 citations · 58 references
ComputingComputer ScienceMachine LearningOptimizationStatistical LearningHilbert Space FormalismKernel MethodsMetric GeometryPattern RecognitionRepresentation LearningAlignment ProblemCorrelationCovarianceDomain AdaptationInfinite-dimensional Hilbert SpacesKernel LearningOptimal TransportReproducing Kernel Hilbert Spaces
In this paper, we present a mathematical and computational framework for comparing and matching distributions in reproducing kernel Hilbert spaces (RKHS). This framework, called optimal transport in RKHS, is a generalization of the optimal transport problem in input spaces to (potentially) infinite-dimensional feature spaces. We provide a computable formulation of Kantorovich's optimal transport in RKHS. In particular, we explore the case in which data distributions in RKHS are Gaussian, obtaining closed-form expressions of both the estimated Wasserstein distance and optimal transport map via kernel matrices. Based on these expressions, we generalize the Bures metric on covariance matrices to infinite-dimensional settings, providing a new metric between covariance operators. Moreover, we extend the correlation alignment problem to Hilbert spaces, giving a new strategy for matching distributions in RKHS. Empirically, we apply the derived formulas under the Gaussianity assumption to image classification and domain adaptation. In both tasks, our algorithms yield state-of-the-art performances, demonstrating the effectiveness and potential of our framework.
58
DeCAF: A Deep Convolutional Activation Feature for Generic Visual Recognition
Jeff Donahue, Yangqing Jia, Oriol Vinyals et al. · arXiv (Cornell University) · 2013 · 3.6K citations · Full text
Sinkhorn Distances: Lightspeed Computation of Optimal Transport
Marco Cuturi · 2013 · 2.2K citations