Optimal and instance-dependent guarantees for Markovian linear stochastic approximation

Wenlong Mou, Ashwin Pananjady, Martin Wainwright, Peter Bartlett
Proceedings of Thirty Fifth Conference on Learning Theory, PMLR 178:2060-2061, 2022.

Abstract

We study stochastic approximation procedures for approximately solving a $d$-dimensional linear fixed point equation based on observing a trajectory of length $n$ from an ergodic Markov chain. We first exhibit a non-asymptotic bound of the order $t_{\mathrm{mix}} \tfrac{d}{n}$ on the squared error of the last iterate of a standard scheme, where $t_{\mathrm{mix}}$ is a mixing time. We then prove a non-asymptotic instance-dependent bound on a suitably averaged sequence of iterates, with a leading term that matches the local asymptotic minimax limit, including sharp dependence on the parameters $(d, t_{\mathrm{mix}})$ in the higher order terms. We complement these upper bounds with a non-asymptotic minimax lower bound that establishes the instance-optimality of the averaged SA estimator. We derive corollaries of these results for policy evaluation with Markov noise—covering the TD($\lambda$) family of algorithms for all $\lambda \in [0, 1)$—and linear autoregressive models. Our instance-dependent characterizations open the door to the design of fine-grained model selection procedures for hyperparameter tuning (e.g., choosing the value of $\lambda$ when running the TD($\lambda$) algorithm).

Cite this Paper

BibTeX
@InProceedings{pmlr-v178-mou22a, title = {Optimal and instance-dependent guarantees for Markovian linear stochastic approximation}, author = {Mou, Wenlong and Pananjady, Ashwin and Wainwright, Martin and Bartlett, Peter}, booktitle = {Proceedings of Thirty Fifth Conference on Learning Theory}, pages = {2060--2061}, year = {2022}, editor = {Loh, Po-Ling and Raginsky, Maxim}, volume = {178}, series = {Proceedings of Machine Learning Research}, month = {02--05 Jul}, publisher = {PMLR}, pdf = {https://proceedings.mlr.press/v178/mou22a/mou22a.pdf}, url = {https://proceedings.mlr.press/v178/mou22a.html}, abstract = {We study stochastic approximation procedures for approximately solving a $d$-dimensional linear fixed point equation based on observing a trajectory of length $n$ from an ergodic Markov chain. We first exhibit a non-asymptotic bound of the order $t_{\mathrm{mix}} \tfrac{d}{n}$ on the squared error of the last iterate of a standard scheme, where $t_{\mathrm{mix}}$ is a mixing time. We then prove a non-asymptotic instance-dependent bound on a suitably averaged sequence of iterates, with a leading term that matches the local asymptotic minimax limit, including sharp dependence on the parameters $(d, t_{\mathrm{mix}})$ in the higher order terms. We complement these upper bounds with a non-asymptotic minimax lower bound that establishes the instance-optimality of the averaged SA estimator. We derive corollaries of these results for policy evaluation with Markov noise—covering the TD($\lambda$) family of algorithms for all $\lambda \in [0, 1)$—and linear autoregressive models. Our instance-dependent characterizations open the door to the design of fine-grained model selection procedures for hyperparameter tuning (e.g., choosing the value of $\lambda$ when running the TD($\lambda$) algorithm).} }
Endnote
%0 Conference Paper %T Optimal and instance-dependent guarantees for Markovian linear stochastic approximation %A Wenlong Mou %A Ashwin Pananjady %A Martin Wainwright %A Peter Bartlett %B Proceedings of Thirty Fifth Conference on Learning Theory %C Proceedings of Machine Learning Research %D 2022 %E Po-Ling Loh %E Maxim Raginsky %F pmlr-v178-mou22a %I PMLR %P 2060--2061 %U https://proceedings.mlr.press/v178/mou22a.html %V 178 %X We study stochastic approximation procedures for approximately solving a $d$-dimensional linear fixed point equation based on observing a trajectory of length $n$ from an ergodic Markov chain. We first exhibit a non-asymptotic bound of the order $t_{\mathrm{mix}} \tfrac{d}{n}$ on the squared error of the last iterate of a standard scheme, where $t_{\mathrm{mix}}$ is a mixing time. We then prove a non-asymptotic instance-dependent bound on a suitably averaged sequence of iterates, with a leading term that matches the local asymptotic minimax limit, including sharp dependence on the parameters $(d, t_{\mathrm{mix}})$ in the higher order terms. We complement these upper bounds with a non-asymptotic minimax lower bound that establishes the instance-optimality of the averaged SA estimator. We derive corollaries of these results for policy evaluation with Markov noise—covering the TD($\lambda$) family of algorithms for all $\lambda \in [0, 1)$—and linear autoregressive models. Our instance-dependent characterizations open the door to the design of fine-grained model selection procedures for hyperparameter tuning (e.g., choosing the value of $\lambda$ when running the TD($\lambda$) algorithm).
APA
Mou, W., Pananjady, A., Wainwright, M. & Bartlett, P.. (2022). Optimal and instance-dependent guarantees for Markovian linear stochastic approximation. Proceedings of Thirty Fifth Conference on Learning Theory, in Proceedings of Machine Learning Research 178:2060-2061 Available from https://proceedings.mlr.press/v178/mou22a.html.

Related Material