Deviation optimal learning using greedy $Q$-aggregation
利用贪心 Q 聚合实现偏差概率意义下的最优学习
Given a finite family of functions, the goal of model selection aggregation is to construct a procedure that mimics the function from this family that is the closest to an unknown regression function. More precisely, we consider a general regression model with fixed design and measure the distance between functions by the mean squared error at the design points. While procedures based on exponential weights are known to solve the problem of model selection aggregation in…