The MADP Toolbox

An Open Source Library for Planning and Learning in (Multi-)Agent Systems

Journal article (2017)

Authors

F.A. Oliehoek University of Liverpool
M.T.J. Spaan Algorithmics - EEMCS
Bas Terwijn Universiteit van Amsterdam
Philipp Robbel Massachusetts Institute of Technology
João V. Messias Universiteit van Amsterdam
Research Group
Algorithmics (EEMCS) (TU Delft)
Copyright: © 2017 F.A. Oliehoek, M.T.J. Spaan, Bas Terwijn, Philipp Robbel, João V. Messias
Software Reinforcement learning Multiagent systems Decision-theoretic planning
More Info
expand_more

Published Date

2017-8

Language

English

Faculty

Electrical Engineering, Mathematics and Computer Science

Department

Software Technology

Research Group

Algorithmics

Abstract

This article describes the MultiAgent Decision Process (MADP) toolbox, a software library to support planning and learning for intelligent agents and multiagent systems in uncertain environments. Key features are that it supports partially observable environments and stochastic transition models; has unified support for single- and multiagent systems; provides a large number of models for decision-theoretic decision making, including one-shot and sequential decision making under various assumptions of observability and cooperation, such as Dec-POMDPs and POSGs; provides tools and parsers to quickly prototype new problems; provides an extensive range of planning and learning algorithms for single- and multiagent systems; it is released under a GNU GPL v3 license; and is written in C++ and designed to be extensible via the object-oriented paradigm.