Date of Award
Spring 2019
Project Type
Thesis
Program or Major
Computer Science
Degree Name
Master of Science
First Advisor
Marek Petrik
Second Advisor
Wheeler Ruml
Third Advisor
Laura Dietz
Abstract
Solving reinforcement learning problems using value function approximation requires having good state features, but constructing them manually is often difficult or impossible. We propose Fast Feature Selection (FFS), a new method for automatically constructing good features in problems with high-dimensional state spaces but low-rank dynamics. Such problems are common when, for example, controlling simple dynamic systems using direct visual observations with states represented by raw images. FFS relies on domain samples and singular value decomposition to construct features that can be used to approximate the optimal value function well. Compared with earlier methods, such as LFD, FFS is simpler and enjoys better theoretical performance guarantees. Our experimental results show that our approach is also more stable, computes better solutions, and can be faster when compared with prior work.
Recommended Citation
Behzadian, Bahram, "Feature Selection by Singular Value Decomposition for Reinforcement Learning" (2019). Master's Theses and Capstones. 1267.
https://scholars.unh.edu/thesis/1267