Fast rates in structured prediction

by   Vivien Cabannes, et al.

Discrete supervised learning problems such as classification are often tackled by introducing a continuous surrogate problem akin to regression. Bounding the original error, between estimate and solution, by the surrogate error endows discrete problems with convergence rates already shown for continuous instances. Yet, current approaches do not leverage the fact that discrete problems are essentially predicting a discrete output when continuous problems are predicting a continuous value. In this paper, we tackle this issue for general structured prediction problems, opening the way to "super fast" rates, that is, convergence rates for the excess risk faster than n^-1, where n is the number of observations, with even exponential rates with the strongest assumptions. We first illustrate it for predictors based on nearest neighbors, generalizing rates known for binary classification to any discrete problem within the framework of structured prediction. We then consider kernel ridge regression where we improve known rates in n^-1/4 to arbitrarily fast rates, depending on a parameter characterizing the hardness of the problem, thus allowing, under smoothness assumptions, to bypass the curse of dimensionality.


page 1

page 2

page 3

page 4


Fast Rates of ERM and Stochastic Approximation: Adaptive to Error Bound Conditions

Error bound conditions (EBC) are properties that characterize the growth...

A Case of Exponential Convergence Rates for SVM

Classification is often the first problem described in introductory mach...

Convergence rates of Kernel Conjugate Gradient for random design regression

We prove statistical rates of convergence for kernel-based least squares...

Regularized Contextual Bandits

We consider the stochastic contextual bandit problem with additional reg...

Error Rates for Kernel Classification under Source and Capacity Conditions

In this manuscript, we consider the problem of kernel classification und...

Learning to Predict Combinatorial Structures

The major challenge in designing a discriminative learning algorithm for...

Asymptotics of Ridge(less) Regression under General Source Condition

We analyze the prediction performance of ridge and ridgeless regression ...