Measuring and Influencing Sequential Joint Agent Behaviours

Measuring and Influencing Sequential Joint Agent Behaviours

Algorithmically designed reward functions can influence groups of learning agents toward measurable desired sequential joint behaviours. Influencing learning agents toward desirable behaviours is non-trivial due to the difficulties of assigning credit for global success to the deserving agents and o...

Full description

Bibliographic Details
Main Author: Raffensperger, Peter Abraham
Language:en
Published: University of Canterbury. Electrical and Computer Engineering 2013
Subjects:
multi-agent systems
multi-agent reinforcement learning
decentralised systems
resource allocation
turn-taking
Nash equilibria
emergent behaviour
reward functions
reward design
Markov decision processes
Markov chains
Online Access:http://hdl.handle.net/10092/7472

Internet

http://hdl.handle.net/10092/7472

Similar Items