Robbins-Monro setting
时间: 2024-05-31 11:08:28 浏览: 49
The Robbins-Monro setting refers to a class of stochastic approximation algorithms used for solving optimization problems. In this setting, the algorithm aims to find the minimum of an unknown objective function by iteratively updating an estimate of the minimum based on noisy observations of the function.
The Robbins-Monro algorithm is particularly useful in situations where the objective function is expensive to evaluate, and the noise in the observations is unpredictable. The algorithm works by adjusting the step size of the updates based on the history of the observations, in order to balance the trade-off between exploration and exploitation.
The Robbins-Monro setting is named after Herbert Robbins and Sutton Monro, who introduced the algorithm in their seminal 1951 paper "A Stochastic Approximation Method". The method has since been widely used in various fields, including machine learning, statistics, and control theory.