@inproceedings{3bcae81a8e234f1fb7dd1d3a918212f6,
title = "Sparse gradient-based direct policy search",
abstract = "Reinforcement learning is challenging if state and action spaces are continuous. The discretization of state and action spaces and real-time adaptation of the discretization are critical issues in reinforcement learning problems. In our contribution we consider the adaptive discretization, and introduce a sparse gradient-based direct policy search method. We address the issue of efficient states/actions selection in the gradient-based direct policy search based on imposing sparsity through the L 1 penalty term. We propose to start learning with a fine discretization of state space and to induce sparsity via the L 1 norm. We compare the proposed approach to state-of-the art methods, such as progressive widening Q-learning which updates the discretization of the states adaptively, and to classic as well as sparse Q-learning with linear function approximation. We demonstrate by our experiments on standard reinforcement learning challenges that the proposed approach is efficient.",
keywords = "Direct policy search, model selection, Q-learning",
author = "Nataliya Sokolovska",
year = "2012",
doi = "10.1007/978-3-642-34478-7_27",
language = "English",
isbn = "9783642344770",
volume = "4",
series = "Lecture notes in computer science",
publisher = "Springer, Springer Nature",
pages = "212--221",
editor = "Tingwen Huang and Zhigang Zeng and Chuandong Li and Leung, {Chi Sing}",
booktitle = "Neural information processing",
address = "United States",
note = "International Conference on Neural Information Processing (19th : 2012) ; Conference date: 12-11-2012 Through 15-11-2012",
}