اشترك بالحزمة الذهبية واحصل على وصول غير محدود شمرا أكاديميا
تسجيل مستخدم جديدNon-Asymptotic Sequential Tests for Overlapping Hypotheses and application to near optimal arm identification in bandit models
75
0
0.0
(
0
)
تأليف
Aurelien Garivier
اسأل ChatGPT حول البحث
ﻻ يوجد ملخص باللغة العربية
In this paper, we study sequential testing problems with emph{overlapping} hypotheses. We first focus on the simple problem of assessing if the mean $mu$ of a Gaussian distribution is $geq -epsilon$ or $leq epsilon$; if $muin(-epsilon,epsilon)$, both answers are considered to be correct. Then, we consider PAC-best arm identification in a bandit model: given $K$ probability distributions on $mathbb{R}$ with means $mu_1,dots,mu_K$, we derive the asymptotic complexity of identifying, with risk at most $delta$, an index $Iin{1,dots,K}$ such that $mu_Igeq max_imu_i -epsilon$. We provide non asymptotic bounds on the error of a parallel General Likelihood Ratio Test, which can also be used for more general testing problems. We further propose lower bound on the number of observation needed to identify a correct hypothesis. Those lower bounds rely on information-theoretic arguments, and specifically on t
قيم البحث
اقرأ أيضاً
This paper considers the change-point problem for finite sequences of networks. To avoid the difficulty of computing the normalization coefficient, such as in Exponential random graphical models (ERGMs) and Markov networks, we construct a finite meas
ure space with measure ratio statistics. A new performance measure of detection delay is proposed to detect the changes in distribution of the network. And an optimal sequential test is proposed under the performance measure. The good performance of the optimal sequential test is illustrated numerically on ERGMs and Erdos-R{e}nyi network sequences.
We present a geometrical method for analyzing sequential estimating procedures. It is based on the design principle of the second-order efficient sequential estimation provided in Okamoto, Amari and Takeuchi (1991). By introducing a dual conformal cu
rvature quantity, we clarify the conditions for the covariance minimization of sequential estimators. These conditions are further elabolated for the multidimensional curved exponential family. The theoretical results are then numerically examined by using typical statistical models, von Mises-Fisher and hyperboloid models.
For some variants of regression models, including partial, measurement error or error-in-variables, latent effects, semi-parametric and otherwise corrupted linear models, the classical parametric tests generally do not perform well. Various modificat
ions and generalizations considered extensively in the literature rests on stringent regularity assumptions which are not likely to be tenable in many applications. However, in such non-standard cases, rank based tests can be adapted better, and further, incorporation of rank analysis of covariance tools enhance their power-efficiency. Numerical studies and a real data illustration show the superiority of rank based inference in such corrupted linear models.
We give a complete characterization of the complexity of best-arm identification in one-parameter bandit problems. We prove a new, tight lower bound on the sample complexity. We propose the `Track-and-Stop strategy, which we prove to be asymptoticall
y optimal. It consists in a new sampling rule (which tracks the optimal proportions of arm draws highlighted by the lower bound) and in a stopping rule named after Chernoff, for which we give a new analysis.
We introduce estimation and test procedures through divergence minimization for models satisfying linear constraints with unknown parameter. Several statistical examples and motivations are given. These procedures extend the empirical likelihood (EL)
method and share common features with generalized empirical likelihood (GEL). We treat the problems of existence and characterization of the divergence projections of probability measures on sets of signed finite measures. Our approach allows for a study of the estimates under misspecification. The asymptotic behavior of the proposed estimates are studied using the dual representation of the divergences and the explicit forms of the divergence projections. We discuss the problem of the choice of the divergence under various respects. Also we handle efficiency and robustness properties of minimum divergence estimates. A simulation study shows that the Hellinger divergence enjoys good efficiency and robustness properties.
سجل دخول لتتمكن من نشر تعليقات
التعليقات
جاري جلب التعليقات
سجل دخول لتتمكن من متابعة معايير البحث التي قمت باختيارها
