TY - JOUR AU - Behzadian, Bahram AU - Gharatappeh, Soheil AU - Petrik, Marek PY - 2021/05/25 Y2 - 2024/03/28 TI - Fast Feature Selection for Linear Value Function Approximation JF - Proceedings of the International Conference on Automated Planning and Scheduling JA - ICAPS VL - 29 IS - 1 SE - Planning and Learning DO - 10.1609/icaps.v29i1.3527 UR - https://ojs.aaai.org/index.php/ICAPS/article/view/3527 SP - 601-609 AB - <p>Linear value function approximation is a standard approach to solving reinforcement learning problems with large state spaces. Since designing good approximation features is difficult, automatic feature selection is an important research topic. We propose a new method for feature selection that is based on a low-rank factorization of the transition matrix. Our approach derives features directly from high-dimensional raw inputs, such as image data. The method is easy to implement using SVD, and our experiments show that it is faster and more stable than alternative methods.</p> ER -