Learning to act using real-time dynamic programming

Artificial Intelligence

Volumn 72, Issue 1-2, 1995, Pages 81-138

  Barto, Andrew G ^a   Bradtke, Steven J ^a   Singh, Satinder P ^a  

^a UNIVERSITY OF MASSACHUSETTS   (United States)

Author keywords

[No Author keywords available]

Indexed keywords

ALGORITHMS; ARTIFICIAL INTELLIGENCE; CONTROL THEORY; DYNAMIC PROGRAMMING; REAL TIME SYSTEMS;

Q-LEARNING ALGORITHM;

LEARNING SYSTEMS;

EID: 0029210635     PISSN: 00043702     EISSN: None     Source Type: Journal    
DOI: 10.1016/0004-3702(94)00011-O     Document Type: Article

References (98)

1. Strategy learning with multilayer connectionist representations
2. Strategy Learning with Multilayer Connectionist Representations
3. Reinforcement learning and adaptive critic methods
4. On the computational economics of reinforcement learning
5. Neuronlike elements that can solve difficult learning control problems
6. Sequential decision problems and neural networks
7. Learning and sequential decision making
8. Functional approximations and dynamic programming
9. Polynomial approximation—a new computational technique in dynamic programming: allocation processes
10. Distributed dynamic programming
11. Reinforcement learning applied to linear quadratic regulation
12. Penquins can make cake
13. Input generalization in delayed reinforcement learning: an algorithm and performance comparisons
14. A unified theory of heuristic evaluation functions and its application to learning
15. A theoretical comparison of the efficiencies of two classical methods and a Monte Carlo method for computing one component of the solution of a set of linear algebraic equations
16. Splines and efficiency in dynamic programming
17. Note on learning rate schedule for stochastic optimization
18. Navigating through temporal difference
19. Reinforcing connectionism: learning the statistical way
20. The convergence of TD(λ) for general λ
21. Contraction mappings in the theory underlying dynamic programming
22. Mathematical games
23. *
24. Universal planning: an (almost) universally bad idea
25. A formal basis for the heuristic determination of minimum cost paths
26. Escaping brittleness: the possibility of general-purpose learning algorithms applied to rule-based systems
27. Computationally efficient adaptive control algorithms for Markov chains
28. Learning to control an unstable system with forward modeling
29. Optimization by simulated annealing
30. Brain function and adaptive systems—a heterostatic theory
31. Proceedings International Conference on Systems, Man, and Cybernetics
32. Real-time heuristic search
33. A survey of some results in stochastic adaptive control
34. The CDP: a unifying formulation for heuristic search, dynamic programming, and branch-and-bound
35. A modified quadratic cost problem and feedback stabilization of a linear system
36. A theoretical framework for back-propagation
37. Real-time optimal path planning using a distributed computing paradigm
38. Programming robots using reinforcement learning and teaching
39. Self-improvement based on reinforcement learning, planning and teaching
40. Self-improving reactive agents: case studies of reinforcement learning frameworks
41. Self-improving reactive agents based on reinforcement learning, planning and teaching
42. Automatic programming of behavior-based robots using reinforcement learning
43. Receding horizon control of nonlinear systems
44. A heuristic search algorithm with modifiable estimate
45. BOXES: an experiment in adaptive control
46. Theory of neural-analog reinforcement systems and its application to the brain-model problem
47. Steps toward artificial intelligence
48. Efficient memory-based learning for robot control
49. Variable resolution dynamic programming: efficiently learning action maps in multivariate real-valued state-spaces
50. Memory-based reinforcement learning: efficient computation with prioritized sweeping
51. Efficient learning and planning within the dyna framework
52. Modified policy iteration algorithms for discounted Markov decision problems
53. Some studies in machine learning using the game of checkers
54. Some studies in machine learning using the game of checkers II—Recent progress
55. Adaptive confidence and adaptive curiosity
56. Universal plans for reactive robots in unpredictable environments
57. In defense of reaction plans as caches
58. An upper bound on the loss from approximate optimal value functions. technical note
59. Temporal credit assignment in reinforcement learning
60. Learning to predict by the method of temporal differences
61. Integrated architectures for learning, planning, and reacting based on approximating dynamic programming
62. Planning by incremental dynamic programming
63. A Special Issue of Machine Learning on Reinforcement Learning
64. Toward a modern theory of adaptive networks: expectation and prediction
65. A temporal-difference model of classical conditioning
66. Time-derivative models of pavlovian reinforcement
67. Reinforcement learning is direct adaptive optimal control
68. Learning a cost-sensitive internal representation for reinforcement learning
69. Practical issues in temporal difference learning
70. The role of exploration in learning control
71. Active exploration in dynamic environments
72. Two kinds of training information for evaluation function learning
73. Learning from delayed rewards
74. Q-learning
75. Approximate dynamic programming for real-time control and neural modeling
76. Beyond regression: new tools for prediction and analysis in the behavioral sciences
77. Advanced forecasting methods for global crisis warning and models of intelligence
78. Applications of advances in nonlinear sensitivity analysis
79. Building and understanding adaptive systems: a statistical/numerical approach to factory automation and brain research
80. Generalization of back propagation with applications to a recurrent gas market model
81. Optimal control: a foundation for intelligent control
82. Complexity and cooperation in Q-learning
83. A mathematical analysis of actor-critic architectures for learning optimal controls through incremental dynamic programming
84. An adaptive optimal controller for discrete-time Markov environments
85. Exploring, modelling and controlling discrete sequential environments
86. Abstraction in control learning