Generalized projections for non-negative functions

DOI10.1007/BF01874442zbMath0837.62006MaRDI QIDQ1897832

Imre Csiszár

Publication date: 20 May 1996

Published in: Acta Mathematica Hungarica (Search for Journal in Brave)

zbMATH Keywords

convex subsets; linear inequalities; generalized projections; \(f\)-divergence; Bergman distance; strictly convex function; \(I\)-projections; existence of projections; generalization of Pythagoras' theorem; measurable nonnegative functions

Mathematics Subject Classification ID

62B10: Statistical aspects of information-theoretic topics

62B99: Sufficiency and information

Related Items

Inference in ϕ-families of distributions, Moment closure approximations of the Boltzmann equation based on \(\varphi \)-divergences, Bayesian model diagnostics using functional Bregman divergence, Generalized projections onto convex sets, Testing for non-nested conditional moment restrictions using unconditional empirical likelihood, Quantization and clustering with Bregman divergences, Solving Fredholm equations by maximum entropy on the mean. Application to superresolution, Projection theorems and estimating equations for power-law models, Game theoretical optimization inspired by information theory, EXISTENCE AND CHARACTERIZATION OF CONDITIONAL DENSITY PROJECTIONS, Robust Statistical Engineering by Means of Scaled Bregman Distances, TESTING FOR NONNESTED CONDITIONAL MOMENT RESTRICTIONS VIA CONDITIONAL EMPIRICAL LIKELIHOOD, Entropic Projections and Dominating Points, Generalized isotone projection cones, Relative entropy and spectral constraints: some invariance properties of the ARMA class

Cites Work

• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Unnamed Item
• Sanov property, generalized I-projection and a conditional limit theorem
• Entropic means
• Why least squares and maximum entropy? An axiomatic approach to inference for linear inverse problems
• I-divergence geometry of probability distributions and minimization problems
• General entropy criteria for inverse problems, with applications to data compression, pattern classification, and cluster analysis
• Approximation–Theoretic Derivation of Logarithmic Entropy Principles for Inverse Problems and Unique Extension of the Maximum Entropy Method to Incorporate Prior Knowledge
• Convergence of best phi -entropy estimates
• Partially-Finite Programming in $L_1 $ and the Existence of Maximum Entropy Estimates