Log In Sign Up

Krylov-Bellman boosting: Super-linear policy evaluation in general state spaces

by   Eric Xia, et al.

We present and analyze the Krylov-Bellman Boosting (KBB) algorithm for policy evaluation in general state spaces. It alternates between fitting the Bellman residual using non-parametric regression (as in boosting), and estimating the value function via the least-squares temporal difference (LSTD) procedure applied with a feature set that grows adaptively over time. By exploiting the connection to Krylov methods, we equip this method with two attractive guarantees. First, we provide a general convergence bound that allows for separate estimation errors in residual fitting and LSTD computation. Consistent with our numerical experiments, this bound shows that convergence rates depend on the restricted spectral structure, and are typically super-linear. Second, by combining this meta-result with sample-size dependent guarantees for residual fitting and LSTD computation, we obtain concrete statistical guarantees that depend on the sample size along with the complexity of the function class used to fit the residuals. We illustrate the behavior of the KBB algorithm for various types of policy evaluation problems, and typically find large reductions in sample complexity relative to the standard approach of fitted value iterationn.


page 1

page 2

page 3

page 4


Variance-Aware Off-Policy Evaluation with Linear Function Approximation

We study the off-policy evaluation (OPE) problem in reinforcement learni...

On the Convergence Rates of Policy Gradient Methods

We consider infinite-horizon discounted Markov decision problems with fi...

Recursive Bias Estimation and L_2 Boosting

This paper presents a general iterative bias correction procedure for re...

A Boosting Approach to Reinforcement Learning

We study efficient algorithms for reinforcement learning in Markov decis...

Incremental Truncated LSTD

Balancing between computational efficiency and sample efficiency is an i...

Stochastic approximation for speeding up LSTD (and LSPI)

We propose a stochastic approximation (SA) based method with randomizati...

Stable Conformal Prediction Sets

When one observes a sequence of variables (x_1, y_1), ..., (x_n, y_n), c...