Dealing With Groups of Actions in Multiagent Markov Decision Processes

Multiagent Markov Decision Processes (MMDPs) provide a useful framework for multiagent decision making. Finding solutions to large-scale problems or with a large number of agents however, has been proven to be computationally hard. In this paper, we adapt H-(PO)MDPs to multi-agent settings by proposing a new approach using action groups to decompose an initial MMDP into a set of dependent Sub-MMDPs where each action group is assigned a corresponding Sub-MMDP. Sub-MMDPs are then solved using a parallel Bellman backup to derive local policies which are synchronized by propagating local results and updating the value functions locally and globally to take the dependencies into account. This decomposition allows, for example, specific aggregation for each sub-MMDP, which we adapt by using a novel value function update. Experimental evaluations have been developed and applied to real robotic platforms showing promising results and validating our techniques.

• Publication year

  2016

• Venue

  International Joint Conference on Computational Intelligence

• Publication date

  Unknown publication date

• Fields of study

  Computer Science

• Identifiers
  DOI 10.5220/0006048000490058
• External record

  Open on Semantic Scholar

• Source metadata

  Semantic Scholar

• No claims are published for this paper.

• No concepts are published for this paper.

• Effective Approximations for Multi-Robot Coordination in Spatially Distributed Tasks

  2015cited by this paper
• Online global loop closure detection for large-scale multi-session graph-based SLAM

  2014cited by this paper
• GSMDPs for Multi-Robot Sequential Decision-Making

  2013cited by this paper
• Coordinated Multi-Robot Exploration Under Communication Constraints Using Decentralized Markov Decision Processes

  2012cited by this paper
• Influence-Based Policy Abstraction for Weakly-Coupled Dec-POMDPs

  2010cited by this paper
• ROS: an open-source Robot Operating System

  2009cited by this paper
• Learning of coordination: exploiting sparse interactions in multiagent systems

  2009cited by this paper
• A Hierarchical Approach to POMDP Planning and Execution

  2004influential reference
• Solving Transition Independent Decentralized Markov Decision Processes

  2004cited by this paper
• Decentralized Control of Cooperative Systems: Categorization and Complexity Analysis

  2004cited by this paper
• Context-specific multiagent coordination and planning with factored MDPs

  2002cited by this paper
• Communication decisions in multi-agent cooperation: model and experiments

  2001cited by this paper
• The Complexity of Decentralized Control of Markov Decision Processes

  2000cited by this paper
• Sequential Optimality and Coordination in Multiagent Systems

  1999cited by this paper
• Flexible Decomposition Algorithms for Weakly Coupled Markov Decision Problems

  1998cited by this paper
• Planning, Learning and Coordination in Multiagent Decision Processes

  1996cited by this paper
• On the Complexity of Solving Markov Decision Problems

  1995cited by this paper
• The Complexity of Markov Decision Processes

  1987cited by this paper
• On the Theory of Dynamic Programming.

  1952cited by this paper

• No citing papers are available for this paper.