@inproceedings{dai2018sbeed,
author = {Dai, Bo and Shaw, Albert and Li, Lihong and Xiao, Lin and He, Niao and Liu, Zhen and Chen, Jianshu and Song, Le},
title = {SBEED: Convergent Reinforcement Learning with Nonlinear Function Approximation},
booktitle = {Proceedings of International Conference on Machine Learning (ICML), 2018},
year = {2018},
month = {July},
abstract = {When function approximation is used, solving the Bellman optimality equation with stability guarantees has remained a major open problem in reinforcement learning for decades. The fundamental difficulty is that the Bellman operator may become an expansion in general, resulting in oscillating and even divergent behavior of popular algorithms like Q-learning. In this paper, we revisit the Bellman equation, and reformulate it into a novel primal-dual optimization problem using Nesterov’s smoothing technique and the Legendre-Fenchel transformation. We then develop a new algorithm, called Smoothed Bellman Error Embedding, to solve this optimization problem where any differentiable function class may be used. We provide what we believe to be the first convergence guarantee for general nonlinear function approximation, and analyze the algorithm’s sample complexity. Empirically, our algorithm compares favorably to state-of-the-art baselines in several benchmark control problems.},
url = {http://approjects.co.za/?big=en-us/research/publication/sbeed-convergent-reinforcement-learning-with-nonlinear-function-approximation/},
journal = {Proceedings of Machine Learning Research},
volume = {80},
}