## The Annals of Probability

### Gaussian mixtures: Entropy and geometric inequalities

#### Abstract

A symmetric random variable is called a Gaussian mixture if it has the same distribution as the product of two independent random variables, one being positive and the other a standard Gaussian random variable. Examples of Gaussian mixtures include random variables with densities proportional to $e^{-|t|^{p}}$ and symmetric $p$-stable random variables, where $p\in(0,2]$. We obtain various sharp moment and entropy comparison estimates for weighted sums of independent Gaussian mixtures and investigate extensions of the B-inequality and the Gaussian correlation inequality in the context of Gaussian mixtures. We also obtain a correlation inequality for symmetric geodesically convex sets in the unit sphere equipped with the normalized surface area measure. We then apply these results to derive sharp constants in Khinchine inequalities for vectors uniformly distributed on the unit balls with respect to $p$-norms and provide short proofs to new and old comparison estimates for geometric parameters of sections and projections of such balls.

#### Article information

Source
Ann. Probab., Volume 46, Number 5 (2018), 2908-2945.

Dates
Revised: October 2017
First available in Project Euclid: 24 August 2018

https://projecteuclid.org/euclid.aop/1535097642

Digital Object Identifier
doi:10.1214/17-AOP1242

Mathematical Reviews number (MathSciNet)
MR3846841

Zentralblatt MATH identifier
06964351

#### Citation

Eskenazis, Alexandros; Nayar, Piotr; Tkocz, Tomasz. Gaussian mixtures: Entropy and geometric inequalities. Ann. Probab. 46 (2018), no. 5, 2908--2945. doi:10.1214/17-AOP1242. https://projecteuclid.org/euclid.aop/1535097642

#### References

• [1] Arora, S. and Kannan, R. (2001). Learning mixtures of arbitrary Gaussians. In Proceedings of the Thirty-Third Annual ACM Symposium on Theory of Computing 247–257. ACM, New York.
• [2] Artstein, S., Ball, K. M., Barthe, F. and Naor, A. (2004). Solution of Shannon’s problem on the monotonicity of entropy. J. Amer. Math. Soc. 17 975–982.
• [3] Artstein-Avidan, S., Giannopoulos, A. and Milman, V. D. (2015). Asymptotic Geometric Analysis, Part I. Mathematical Surveys and Monographs 202. Amer. Math. Soc., Providence, RI.
• [4] Averkamp, R. and Houdré, C. (2003). Wavelet thresholding for non-necessarily Gaussian noise: Idealism. Ann. Statist. 31 110–151.
• [5] Baernstein, A. II and Culverhouse, R. C. (2002). Majorization of sequences, sharp vector Khinchin inequalities, and bisubharmonic functions. Studia Math. 152 231–248.
• [6] Ball, K. (1986). Cube slicing in $\textbf{R}^{n}$. Proc. Amer. Math. Soc. 97 465–473.
• [7] Ball, K., Nayar, P. and Tkocz, T. (2016). A reverse entropy power inequality for log-concave random vectors. Studia Math. 235 17–30.
• [8] Barthe, F. (1995). Mesures unimodales et sections des boules $B^{n}_{p}$. C. R. Acad. Sci. Paris Sér. I Math. 321 865–868.
• [9] Barthe, F., Guédon, O., Mendelson, S. and Naor, A. (2005). A probabilistic approach to the geometry of the $l^{n}_{p}$-ball. Ann. Probab. 33 480–513.
• [10] Barthe, F. and Naor, A. (2002). Hyperplane projections of the unit ball of $l^{n}_{p}$. Discrete Comput. Geom. 27 215–226.
• [11] Bobkov, S. G. and Chistyakov, G. P. (2015). Entropy power inequality for the Rényi entropy. IEEE Trans. Inform. Theory 61 708–714.
• [12] Bobkov, S. G. and Houdré, C. (1996). Characterization of Gaussian measures in terms of the isoperimetric property of half-spaces. Zap. Nauchn. Sem. S.-Peterburg. Otdel. Mat. Inst. Steklov. (POMI) 228 31–38, 356.
• [13] Bobkov, S. G. and Houdré, C. (1997). Isoperimetric constants for product probability measures. Ann. Probab. 25 184–205.
• [14] Böröczky, K. J., Lutwak, E., Yang, D. and Zhang, G. (2012). The log-Brunn–Minkowski inequality. Adv. Math. 231 1974–1997.
• [15] Cordero-Erausquin, D., Fradelizi, M. and Maurey, B. (2004). The (B) conjecture for the Gaussian measure of dilates of symmetric convex sets and related problems. J. Funct. Anal. 214 410–427.
• [16] Dasgupta, S. (1999). Learning mixtures of Gaussians. In 40th Annual Symposium on Foundations of Computer Science (New York, 1999) 634–644. IEEE Computer Soc., Los Alamitos, CA.
• [17] Eskenazis, A. and Nayar, P. and Tkocz, T. (2018). Sharp comparison of moments and the log-concave moment problem. Preprint. Available at arXiv:1801.07597..
• [18] Feller, W. (1971). An Introduction to Probability Theory and Its Applications, Vol. II. 2nd ed. Wiley, New York.
• [19] Gozlan, N. and Léonard, C. (2010). Transport inequalities. A survey. Markov Process. Related Fields 16 635–736.
• [20] Guédon, O., Nayar, P. and Tkocz, T. (2014). Concentration inequalities and geometry of convex bodies. In Analytical and Probabilistic Methods in the Geometry of Convex Bodies. IMPAN Lect. Notes 2 9–86. Polish Acad. Sci. Inst. Math., Warsaw.
• [21] Haagerup, U. (1981). The best constants in the Khintchine inequality. Studia Math. 70 231–283 (1982).
• [22] Hardy, G. H., Littlewood, J. E. and Pólya, G. (1988). Inequalities. Cambridge Univ. Press, Cambridge. Reprint of the 1952 edition.
• [23] Klartag, B. and Vershynin, R. (2007). Small ball probability and Dvoretzky’s theorem. Israel J. Math. 157 193–207.
• [24] Koldobsky, A. (1998). An application of the Fourier transform to sections of star bodies. Israel J. Math. 106 157–164.
• [25] Koldobsky, A. (2005). Fourier Analysis in Convex Geometry. Mathematical Surveys and Monographs 116. Amer. Math. Soc., Providence, RI.
• [26] Koldobsky, A. and Zymonopoulou, M. (2003). Extremal sections of complex $l_{p}$-balls, $0<p\le2$. Studia Math. 159 185–194.
• [27] Koldobsky, A. L. and Montgomery-Smith, S. J. (1996). Inequalities of correlation type for symmetric stable random vectors. Statist. Probab. Lett. 28 91–97.
• [28] König, H. (2014). On the best constants in the Khintchine inequality for Steinhaus variables. Israel J. Math. 203 23–57.
• [29] Latała, R. (2002). On some inequalities for Gaussian measures. In Proceedings of the International Congress of Mathematicians, Vol. II (Beijing, 2002) 813–822. Higher Ed. Press, Beijing.
• [30] Latała, R. and Matlak, D. (2017). Royen’s proof of the Gaussian correlation inequality. In Geometric Aspects of Functional Analysis. Lecture Notes in Math. 2169 265–275. Springer, Cham.
• [31] Latała, R. and Oleszkiewicz, K. (1995). A note on sums of independent uniformly distributed random variables. Colloq. Math. 68 197–206.
• [32] Latała, R. and Oleszkiewicz, K. (2005). Small ball probability estimates in terms of widths. Studia Math. 169 305–314.
• [33] Lewis, T. M. and Pritchard, G. (1999). Correlation measures. Electron. Commun. Probab. 4 77–85.
• [34] Lewis, T. M. and Pritchard, G. (2003). Tail properties of correlation measures. J. Theoret. Probab. 16 771–788.
• [35] Lieb, E. H. (1978). Proof of an entropy conjecture of Wehrl. Comm. Math. Phys. 62 35–41.
• [36] Livne Bar-on, A. (2014). The (B) conjecture for uniform measures in the plane. In Geometric Aspects of Functional Analysis. Lecture Notes in Math. 2116 341–353. Springer, Cham.
• [37] Marshall, A. W. and Olkin, I. (1979). Inequalities: Theory of Majorization and Its Applications. Mathematics in Science and Engineering 143. Academic Press, New York.
• [38] Marshall, A. W. and Proschan, F. (1965). An inequality for convex functions involving majorization. J. Math. Anal. Appl. 12 87–90.
• [39] Marsiglietti, A. (2016). On the improvement of concavity of convex measures. Proc. Amer. Math. Soc. 144 775–786.
• [40] Memarian, Y. (2015). On a correlation inequality for Cauchy type measures. New Zealand J. Math. 45 53–64.
• [41] Meyer, M. and Pajor, A. (1988). Sections of the unit ball of $l^{n}_{p}$. J. Funct. Anal. 80 109–123.
• [42] Nayar, P. and Oleszkiewicz, K. (2012). Khinchine type inequalities with optimal constants via ultra log-concavity. Positivity 16 359–371.
• [43] Paouris, G. and Valettas, P. (2018). A small deviation inequality for convex functions. Ann. Probab. 46 1441–1454.
• [44] Paouris, G. and Valettas, P. (2016). Variance estimates and almost Euclidean structure. Preprint. Available at arXiv:1703.10244.
• [45] Rachev, S. T. and Rüschendorf, L. (1991). Approximate independence of distributions on spheres and their stability properties. Ann. Probab. 19 1311–1337.
• [46] Royen, T. (2014). A simple proof of the Gaussian correlation conjecture extended to some multivariate gamma distributions. Far East J. Theor. Stat. 48 139–145.
• [47] Saroglou, C. (2015). Remarks on the conjectured log-Brunn–Minkowski inequality. Geom. Dedicata 177 353–365.
• [48] Saroglou, C. (2016). More on logarithmic sums of convex bodies. Mathematika 62 818–841.
• [49] Schechtman, G. and Zinn, J. (1990). On the volume of the intersection of two $L^{n}_{p}$ balls. Proc. Amer. Math. Soc. 110 217–224.
• [50] Shannon, C. E. and Weaver, W. (1949). The Mathematical Theory of Communication. Univ. Illinois Press, Urbana, IL.
• [51] Simon, T. (2011). Multiplicative strong unimodality for positive stable laws. Proc. Amer. Math. Soc. 139 2587–2595.
• [52] Stam, A. J. (1959). Some inequalities satisfied by the quantities of information of Fisher and Shannon. Inf. Control 2 101–112.
• [53] Szarek, S. J. (1976). On the best constants in the Khinchin inequality. Studia Math. 58 197–208.
• [54] Vershynin, R. (2009). Lectures in geometric functional analysis. Lecture notes, available at https://www.math.uci.edu/~rvershyn/papers/GFA-book.pdf.
• [55] Weron, A. (1984). Stable processes and measures: A survey. In Probability Theory on Vector Spaces, III (Lublin, 1983). Lecture Notes in Math. 1080 306–364. Springer, Berlin.
• [56] Whittle, P. (1960). Bounds for the moments of linear and quadratic forms in independent variables. Teor. Verojatnost. i Primenen. 5 331–335.
• [57] Yu, Y. (2008). Letter to the editor: On an inequality of Karlin and Rinott concerning weighted sums of i.i.d. random variables. Adv. in Appl. Probab. 40 1223–1226.