Quantitative Stability of the Entropy Power Inequality
- 20 February 2018
- journal article
- research article
- Published by Institute of Electrical and Electronics Engineers (IEEE) in IEEE Transactions on Information Theory
- Vol. 64 (8) , 5691-5703
Abstract
We establish quantitative stability results for the entropy power inequality (EPI). Specifically, we show that if uniformly log-concave densities nearly saturate the EPI, then they must be close to Gaussian densities in the quadratic Kantorovich-Wasserstein distance. Furthermore, if one of the densities is Gaussian and the other is log-concave, or more generally has positive spectral gap, then the deficit in the EPI can be controlled in terms of the L 1 -Kantorovich-Wasserstein distance or relative entropy, respectively. As a counterpoint, an example shows that the EPI can be unstable with respect to the quadratic Kantorovich-Wasserstein distance when densities are uniformly log-concave on sets of measure arbitrarily close to one. Our stability results can be extended to non-log-concave densities, provided certain regularity conditions are met. The proofs are based on mass transportation.Keywords
Funding Information
- National Science Foundation (CCF-1528132, CCF-0939370)
- National Science Foundation (DMS-1361122)
- France–Berkeley Fund
This publication has 47 references indexed in Scilit:
- Entropy jumps for isotropic log-concave random vectors and spectral gapStudia Mathematica, 2012
- Entropy and chaos in the Kac modelKinetic and Related Models, 2010
- Convergence of Differential EntropiesIEEE Transactions on Information Theory, 2004
- Entropy jumps in the presence of a spectral gapDuke Mathematical Journal, 2003
- Entropy production by block variable summation and central limit theoremsCommunications in Mathematical Physics, 1991
- Entropy and the Central Limit TheoremThe Annals of Probability, 1986
- Proof of an entropy conjecture of WehrlCommunications in Mathematical Physics, 1978
- Logarithmic Sobolev InequalitiesAmerican Journal of Mathematics, 1975
- The convolution inequality for entropy powersIEEE Transactions on Information Theory, 1965
- Contributions to the theory of convex bodies.The Michigan Mathematical Journal, 1957