Kernel-based reinforcement learning

Machine Learning

Volumn 49, Issue 2-3, 2002, Pages 161-178

  Ormoneit, Dirk ^a   Sen, Åśaunak ^b  

^a Stanford University ^*  (United States)

^b JACKSON LABORATORY   (United States)

Author keywords

Kernel smoothing; Kernel based learning; Lazy learning; Local averaging; Markov decision process; Reinforcement learning

Indexed keywords

APPROXIMATION THEORY; ASYMPTOTIC STABILITY; CONVERGENCE OF NUMERICAL METHODS; LEARNING ALGORITHMS; MARKOV PROCESSES; NEURAL NETWORKS; OPTIMIZATION; PARAMETER ESTIMATION; REGRESSION ANALYSIS; STATE SPACE METHODS;

BELLMANS EQUATION; GAUSSIAN PROCESS; KERNEL BASED LEARNING; LAZY LEARNING; MARKOV DECISION PROCESS; REINFORCEMENT LEARNING; TEMPORAL DIFFERENCE LEARNING;

LEARNING SYSTEMS;

EID: 0036832956     PISSN: 08856125     EISSN: None     Source Type: Journal    
DOI: 10.1023/A:1017928328829     Document Type: Article

References (35)

1. Locally weighted regression for control
2. Reinforcement learning with high-dimensional, continuous actions
3. Generalization in reinforcement learning: Safely approximating the value function
4. Reinforcement learning applied to linear quadratic regulation
5. Estimating portfolio and consumption choice. A conditional Euler equations approach
6. Learning to control a dynamic physical system
7. Approximate solutions to Markov decision processes
8. Local regression: Autmatic kernel carpentry
9. Central limit theorems for C(S)-valued random variables
10. Reinforcement learning and distributed local model synthesis
11. Valuing American options by simulations: A simple least-squares approach
12. Kernel-based reinforcement learning in average-cost problems: An application to optimal portfolio choice
13. Kernel-based reinforcement learning in average-cost problems
14. Optimal kernel shapes for local linear regression
15. Convergence of reinforcement learning with general function approximators
16. Efficient learning and planning within the Dyna framework
17. A stochastic approximation method
18. Using randomization to break the curse of dimensionality
19. Reinforcement learning for dynamic channel allocation in cellular telephone systems
20. Practical reinforcement learning in continuous spaces
21. Optimal global rates of convergence for nonparametric regression
22. Learning to predict by the methods of temporal differences
23. Policy gradient methods for reinforcement learning with function approximation
24. Neurogammon wins computer olympiad
25. Active exploration in dynamic environments
26. Feature-based methods for large-scale dynamic programming
27. Optimal stopping of Markov processes: Hilbert space theory approximation algorithms, and an application to pricing high-dimensional financial derivatives
28. Actor-critic algorithms
29. Q-learning
30. Consistency of HDP applied to a simple reinforcement learning problem