Erratum: Concentration bounds for stochastic approximation
نویسندگان
چکیده
منابع مشابه
Tail Bounds for Stochastic Approximation
Stochastic-approximation gradient methods are attractive for large-scale convex optimization because they offer inexpensive iterations. They are especially popular in data-fitting and machine-learning applications where the data arrives in a continuous stream, or it is necessary to minimize large sums of functions. It is known that by appropriately decreasing the variance of the error at each i...
متن کاملConcentration Bounds for Two Timescale Stochastic Approximation with Applications to Reinforcement Learning
Two-timescale Stochastic Approximation (SA) algorithms are widely used in Reinforcement Learning (RL). In such methods, the iterates consist of two parts that are updated using different stepsizes. We develop the first convergence rate result for these algorithms; in particular, we provide a general methodology for analyzing two-timescale linear SA. We apply our methodology to two-timescale RL ...
متن کاملstrong approximation for itô stochastic differential equations
in this paper, a class of semi-implicit two-stage stochastic runge-kutta methods (srks) of strong global order one, with minimum principal error constants are given. these methods are applied to solve itô stochastic differential equations (sdes) with a wiener process. the efficiency of this method with respect to explicit two-stage itô runge-kutta methods (irks), it method, milstien method, sem...
متن کاملBounds on the concentration function in terms of Diophantine approximation
We demonstrate a simple analytic argument that may be used to bound the Lévy concentration function of a sum of independent random variables. The main application is a version of a recent inequality due to Rudelson and Vershynin, and its multidimensional generalisa-tion. Des bornes pour la fonction de concentration enmatì ere d'approximation Diophantienne. Nous montrons un simple raison-nement ...
متن کاملAsymptotic Behaviour of Truncated Stochastic Approximation procedures with Moving Bounds
Asymptotic behaviour of stochastic approximation procedures is studeded with three main characteristics: truncations with random moving bounds, a matrix valued random step-size sequence, and a dynamically changing random regression function. In particular, the rate of convergence and asymptotic linearity are established. Main results are supplemented with corollaries to establish different sets...
متن کاملذخیره در منابع من
با ذخیره ی این منبع در منابع من، دسترسی به آن را برای استفاده های بعدی آسان تر کنید
ژورنال
عنوان ژورنال: Electronic Communications in Probability
سال: 2012
ISSN: 1083-589X
DOI: 10.1214/ecp.v17-2495