Nonparametric Infinite Horizon Kullback-Leibler Stochastic Control
arXiv:1404.3984
Abstract
We present two nonparametric approaches to Kullback-Leibler (KL) control, or linearly-solvable Markov decision problem (LMDP) based on Gaussian processes (GP) and Nyström approximation. Compared to recently developed parametric methods, the proposed data-driven frameworks feature accurate function approximation and efficient on-line operations. Theoretically, we derive the mathematical connection of KL control based on dynamic programming with earlier work in control theory which relies on information theoretic dualities for the infinite time horizon case. Algorithmically, we give explicit optimal control policies in nonparametric forms, and propose on-line update schemes with budgeted computational costs. Numerical results demonstrate the effectiveness and usefulness of the proposed frameworks.
References in corpus (4)
- Path integrals and symmetry breaking for optimal control theory
- A linear theory for control of non-linear stochastic systems
- Coarse-graining the Dynamics of a Driven Interface in the Presence of Mobile Impurities: Effective Description via Diffusion Maps
- Path Integral Control by Reproducing Kernel Hilbert Space Embedding