Abstract
In the previous chapter, we proposed a MAL algorithm CMLeS that in a arbitrary repeated game, converges to following a NE joint-policy in self-play, achieves close to the best response with a high probability against a set of memory-bounded agents whose memory size is upper-bounded by a known value, and achieves close to the security value against any other set of agents which cannot be represented as being K max memory-bounded. CMLeS is the first MAL algorithm to achieve all of the above objectives.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
Preview
Unable to display preview. Download preview PDF.
Author information
Authors and Affiliations
Corresponding author
Rights and permissions
Copyright information
© 2014 Springer International Publishing Switzerland
About this chapter
Cite this chapter
Chakraborty, D. (2014). Maximizing Social Welfare in the Presence of Markovian Agents. In: Sample Efficient Multiagent Learning in the Presence of Markovian Agents. Studies in Computational Intelligence, vol 523. Springer, Cham. https://doi.org/10.1007/978-3-319-02606-0_5
Download citation
DOI: https://doi.org/10.1007/978-3-319-02606-0_5
Publisher Name: Springer, Cham
Print ISBN: 978-3-319-02605-3
Online ISBN: 978-3-319-02606-0
eBook Packages: EngineeringEngineering (R0)