Optimism in Reinforcement Learning with Generalized Linear Function Approximation

Authors: Yining Wang, Ruosong Wang, Simon Shaolei Du, Akshay Krishnamurthy

ICLR 2021 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Theoretical We design a new provably efficient algorithm for episodic reinforcement learning with generalized linear function approximation. We analyze the algorithm under a new expressivity assumption that we call optimistic closure, which is strictly weaker than assumptions from prior analyses for the linear setting. With optimistic closure, we prove that our algorithm enjoys a regret bound of e O H where H is the horizon, d is the dimensionality of the state-action features and T is the number of episodes. This is the first statistically and computationally efficient algorithm for reinforcement learning with generalized linear functions.
Researcher Affiliation Collaboration Yining Wang University of Florida yining.wang@warrington.ufl.edu Ruosong Wang Carnegie Mellon University ruosongw@andrew.cmu.edu Simon S. Du University of Washington ssdu@cs.washington.edu Akshay Krishnamurthy Microsoft Research akshaykr@microsoft.com
Pseudocode Yes Algorithm 1 The LSVI-UCB algorithm with generalized linear function approximation.
Open Source Code No The paper does not provide any information about open-sourcing the code for the described methodology.
Open Datasets No The paper is theoretical and does not use or refer to any specific publicly available datasets for training. It discusses abstract MDPs and function approximation.
Dataset Splits No The paper is theoretical and does not describe any dataset splits for training, validation, or testing.
Hardware Specification No The paper is theoretical and does not describe any specific hardware used for experiments.
Software Dependencies No The paper is theoretical and does not mention specific software dependencies with version numbers.
Experiment Setup No The paper is theoretical and defines mathematical parameters and constants for its analysis, but it does not provide concrete experimental setup details like hyperparameters (e.g., learning rate, batch size) or specific training configurations.