Multiagent learning using a variable learning rate

Artificial Intelligence

Volumn 136, Issue 2, 2002, Pages 215-250

  Bowling, Michael ^a   Veloso, Manuela ^a  

^a Carnegie Mellon University ^*  (United States)

Author keywords

Game theory; Multiagent learning; Reinforcement learning

Indexed keywords

ARTIFICIAL INTELLIGENCE; CONVERGENCE OF NUMERICAL METHODS; GAME THEORY; LEARNING ALGORITHMS; PROBLEM SOLVING; RANDOM PROCESSES;

MULTIAGENT LEARNING; REINFORCEMENT SYSTEMS;

MULTI AGENT SYSTEMS;

EID: 0036531878     PISSN: 00043702     EISSN: None     Source Type: Journal    
DOI: 10.1016/S0004-3702(02)00121-2     Document Type: Article

References (35)

1. Gradient descent for general reinforcement learning
2. Reinforcement learning in pomdp's via direct gradient ascent
3. On-line learning and the metrical task system problem
4. Convergence problems of general-sum multiagent reinforcement learning
5. Rational and convergent learning in stochastic games
6. Variable learning rate and the convergence of gradient dynamics
7. The dynamics of reinforcement learning in cooperative multiagent systems
8. Equilibrium in a stochastic n-person game
9. Multiagent reinforcement learning: Theoretical framework and an algorithm
10. Reinforcement learning algorithm for partially observable Markov decision problems
11. Reinforcement learning: A survey
12. Markov games as a framework for multi-agent reinforcement learning
13. Two-person nonzero-sum games and quadratic programming
14. Equilibrium points in n-person games
15. An iterative method of solving a game
16. Learning to coordinate without sharing information
17. Stochastic games
18. Nash convergence of gradient dynamics in general-sum games
19. Convergence results for single-step on-policy reinforcement-learning algorithms
20. Policy gradient methods for reinforcement learning with function approximation
21. Multi-agent reinforcement learning: Independent vs. cooperative agents