Advances in Soviet Mathematics 1992; 150 pp; hardcover Volume: 12 ISBN10: 0821841114 ISBN13: 9780821841112 List Price: US$120 Member Price: US$96 Order Code: ADVSOV/12
 This book contains papers presented at the Seminar on Mathematical Statistics held at the Institute for Problems of Information Transmission of the Academy of Sciences in the former Soviet Union. Founded in the mid1960s, this seminar is still active today and attracts most of the researchers in Moscow who are interested in mathematical statistics. The topics covered include density, regression, and image estimation, adaptive estimation, stochastic approximation, median estimation, sequential experimental design, and large deviations for empirical measures. This collection is distinguished by the high scientific level of the papers and their modern approach. This book will be of interest to scientists and engineers who use probability and statistics, to mathematicians and applied statisticians who work in approximation theory, and to computer scientists who work in image analysis. Readership Scientists and mathematicians in probability, statistics, and approximation theory. Computer scientists interested in image analysis. Table of Contents  A. Samarov  Lower bound for the integral risk of density function estimates
 A. S. Nemirovskiĭ  On nonparametric estimation of functions satisfying differential inequalities
 A. P. Korostelev and A. B. Tsybakov  Asymptotically minimax image reconstruction problems
 O. V. Lepskiĭ  On problems of adaptive estimation in white Gaussian noise
 B. T. Polyak and A. B. Tsybakov  On stochastic approximation with arbitrary noise (the KWcase)
 E. N. Belitser and A. P. Korostelev  Pseudovalues and minimax filtering algorithms for the nonparametric median
 A. Yu. Veretennikov  On large deviations for ergodic process empirical measures
 V. G. Spokoĭnyĭ  On asymptotically optimal sequential experimental design
