Sciweavers

Free Online Productivity Tools i2Speak i2Symbol i2OCR iTex2Img iWeb2Print iWeb2Shot i2Type iPdf2Split iPdf2Merge i2Bopomofo i2Arabic i2Style i2Image i2PDF iLatex2Rtf Sci2ools

31

ICCCI
2011
Springer

favoriteEmaildiscussreport

223views Intelligent Agents» more ICCCI 2011»

Evolving Equilibrium Policies for a Multiagent Reinforcement Learning Problem with State Attractors

12 years 3 months ago

Evolving Equilibrium Policies for a Multiagent Reinforcement Learning Problem with State Attractors

Download florinleon.byethost24.com

Multiagent reinforcement learning problems are especially difficult because of their dynamism and the size of joint state space. In this paper a new benchmark problem is proposed, which involves the need for cooperation, competition and synchronization between agents. The notion of state attractor is introduced, such that agents compute their actions based on the proximity of their current state to the nearest state attractor. A genetic algorithm is used to find the state attractors. This representation can be used as a compact way to define individual or joint policies.

Florin Leon

Real-time Traffic

Benchmark Problem | Genetic Algorithm | ICCCI 2011 | Intelligent Agents | Reinforcement Learning |

claim paper

Related Content

» A Multiagent Reinforcement Learning Algorithm with Nonlinear Dynamics

» Convergence Problems of GeneralSum Multiagent Reinforcement Learning

» Formalizing Multistate Learning Dynamics

» MultiAgent Learning with Policy Prediction

» Learning of coordination exploiting sparse interactions in multiagent systems

» Decentralized Learning in Markov Games

» Modular Learning System and Scheduling for Behavior Acquisition in Multiagent Environment

» Learning multiagent state space representations

» Reinforcement learning for DECMDPs with changing action sets and partially ordered depende...

Post Info
More Details (n/a)

Added	24 Dec 2011
Updated	24 Dec 2011
Type	Journal
Year	2011
Where	ICCCI
Authors	Florin Leon

Comments (0)