Skip to main content
×
×
Home

Multi-agent credit assignment in stochastic resource management games

  • Patrick Mannion (a1) (a2), Sam Devlin (a3), Jim Duggan (a2) and Enda Howley (a2)
Abstract
Abstract

Multi-agent systems (MASs) are a form of distributed intelligence, where multiple autonomous agents act in a common environment. Numerous complex, real world systems have been successfully optimized using multi-agent reinforcement learning (MARL) in conjunction with the MAS framework. In MARL agents learn by maximizing a scalar reward signal from the environment, and thus the design of the reward function directly affects the policies learned. In this work, we address the issue of appropriate multi-agent credit assignment in stochastic resource management games. We propose two new stochastic games to serve as testbeds for MARL research into resource management problems: the tragic commons domain and the shepherd problem domain. Our empirical work evaluates the performance of two commonly used reward shaping techniques: potential-based reward shaping and difference rewards. Experimental results demonstrate that systems using appropriate reward shaping techniques for multi-agent credit assignment can achieve near-optimal performance in stochastic resource management games, outperforming systems learning using unshaped local or global evaluations. We also present the first empirical investigations into the effect of expressing the same heuristic knowledge in state- or action-based formats, therefore developing insights into the design of multi-agent potential functions that will inform future work.

  • View HTML
    • Send article to Kindle

      To send this article to your Kindle, first ensure no-reply@cambridge.org is added to your Approved Personal Document E-mail List under your Personal Document Settings on the Manage Your Content and Devices page of your Amazon account. Then enter the ‘name’ part of your Kindle email address below. Find out more about sending to your Kindle. Find out more about sending to your Kindle.

      Note you can select to send to either the @free.kindle.com or @kindle.com variations. ‘@free.kindle.com’ emails are free but can only be sent to your device when it is connected to wi-fi. ‘@kindle.com’ emails can be delivered even when you are not connected to wi-fi, but note that service fees apply.

      Find out more about the Kindle Personal Document Service.

      Multi-agent credit assignment in stochastic resource management games
      Available formats
      ×
      Send article to Dropbox

      To send this article to your Dropbox account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your Dropbox account. Find out more about sending content to Dropbox.

      Multi-agent credit assignment in stochastic resource management games
      Available formats
      ×
      Send article to Google Drive

      To send this article to your Google Drive account, please select one or more formats and confirm that you agree to abide by our usage policies. If this is the first time you use this feature, you will be asked to authorise Cambridge Core to connect with your Google Drive account. Find out more about sending content to Google Drive.

      Multi-agent credit assignment in stochastic resource management games
      Available formats
      ×
Copyright
References
Hide All
Arthur W. B. 1994. Inductive reasoning and bounded rationality. The American Economic Review 84, 406411.
Binmore K. 2012. Playing for Real: A Text on Game Theory. Oxford University Press.
Buşoniu L., Babuška R. & Schutter B. 2010. Multi-agent reinforcement learning: An overview. In Innovations in Multi-Agent Systems and Applications – 1, Volume 310 of Studies in Computational Intelligence, Srinivasan, D. & Jain, L. (eds). Springer Berlin Heidelberg, 183221.
Colby M., Duchow-Pressley T., Chung J. J. & Tumer K. 2016. Local approximation of difference evaluation functions. In Proceedings of the 15th International Conference on Autonomous Agents & Multiagent Systems (AAMAS), 521–529.
de Jong S. & Tuyls K. 2009. Learning to cooperate in a continuous tragedy of the commons. In Proceedings of the 8th International Conference on Autonomous Agents and Multiagent Systems – Volume 2, 1185–1186. International Foundation for Autonomous Agents and Multiagent Systems.
Devlin S. & Kudenko D. 2011. Theoretical considerations of potential-based reward shaping for multi-agent systems. In Proceedings of the 10th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 225–232.
Devlin S. & Kudenko D. 2012. Dynamic potential-based reward shaping. In Proceedings of the 11th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 433–440.
Devlin S., Grzes M. & Kudenko D. 2011a. An empirical study of potential-based reward shaping and advice in complex, multi-agent systems. Advances in Complex Systems 14(2), 251278.
Devlin S., Grzes M. & Kudenko D. 2011b. Multi-agent, potential-based reward shaping for robocup keepaway (extended abstract). In Proceedings of the 10th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 1227–1228.
Devlin S., Yliniemi L., Kudenko D. & Tumer K. 2014. Potential-based difference rewards for multiagent reinforcement learning. In Proceedings of the 13th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 165–172.
Grześ M. & Kudenko D. 2009. Theoretical and empirical analysis of reward shaping in reinforcement learning. In International Conference on Machine Learning and Applications, 2009. ICMLA’09, 337–344. IEEE.
Howley E. & Duggan J. 2011. Investing in the commons: a study of openness and the emergence of cooperation. Advances in Complex Systems 14(2), 229250.
Malialis K., Devlin S. & Kudenko D. 2015. Distributed reinforcement learning for adaptive and robust network intrusion response. Connection Science 27(3), 234252.
Malialis K., Devlin S. & Kudenko D. 2016. Resource abstraction for reinforcement learning in multiagent congestion problems. In Proceedings of the 15th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 503–511.
Mannion P., Duggan J. & Howley E. 2016a. An experimental review of reinforcement learning algorithms for adaptive traffic signal control. In Autonomic Road Transport Support Systems, McCluskey, L. T., Kotsialos, A., Müller, P. J., Klügl, F., Rana, O. & Schumann, R. (eds). Springer International Publishing, 4766.
Mannion P., Duggan J. & Howley E. 2016b. Generating multi-agent potential functions using counterfactual estimates. In Proceedings of Learning, Inference and Control of Multi-Agent Systems (at NIPS 2016).
Mannion P., Duggan J. & Howley E. 2017. Analysing the effects of reward shaping in multi-objective stochastic games. In Proceedings of the Adaptive and Learning Agents Workshop (at AAMAS 2017).
Mannion P., Mason K., Devlin S., Duggan J. & Howley E. 2016c. Dynamic economic emissions dispatch optimisation using multi-agent reinforcement learning. In Proceedings of the Adaptive and Learning Agents Workshop (at AAMAS 2016).
Mannion P., Mason K., Devlin S., Duggan J. & Howley E. 2016d. Multi-objective dynamic dispatch optimisation using multi-agent reinforcement learning. In Proceedings of the 15th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 1345–1346.
Mason K., Mannion P., Duggan J. & Howley E. 2016. Applying multi-agent reinforcement learning to watershed management. In Proceedings of the Adaptive and Learning Agents Workshop (at AAMAS 2016).
Ng A. Y., Harada D. & Russell S. J. 1999. Policy invariance under reward transformations: theory and application to reward shaping. In Proceedings of the Sixteenth International Conference on Machine Learning, ICML ’99, 278–287. Morgan Kaufmann Publishers Inc.
Puterman M. L. 1994. Markov Decision Processes: Discrete Stochastic Dynamic Programming, 1st edition. John Wiley & Sons, Inc.
Randløv J. & Alstrøm P. 1998. Learning to drive a bicycle using reinforcement learning and shaping. In Proceedings of the Fifteenth International Conference on Machine Learning, ICML ’98, 463–471. Morgan Kaufmann Publishers Inc.
Shoham Y., Powers R. & Grenager T. 2007. If multi-agent learning is the answer, what is the question? Artificial Intelligence 171, 365377.
Tumer K. & Agogino A. 2007. Distributed agent-based air traffic flow management. In Proceedings of the 6th International Conference on Autonomous Agents and Multiagent Systems (AAMAS), 330–337.
Tumer K., Agogino A. K. & Welch Z. 2009. Traffic congestion management as a learning agent coordination problem. In Multiagent Architectures for Traffic and Transportation Engineering, Bazzan, A. & Kluegl, F. (eds). Lecture notes in AI. Springer, 261279.
Watkins C. J. & Dayan P. 1992. Technical note: Q-learning. Machine Learning 8, 279292.
Watkins C. J. C. H. 1989. Learning from Delayed Rewards. PhD thesis, King’s College.
Wiering M. & van Otterlo M. (eds) 2012. Reinforcement Learning: State-of-the-Art. Springer.
Wiewiora E., Cottrell G. & Elkan C. 2003. Principled methods for advising reinforcement learning agents. In Proceedings of the Twentieth International Conference on Machine Learning, 792–799.
Wolpert D. H. & Tumer K. 2002. Collective intelligence, data routing and Braess’ paradox. Journal of Artificial Intelligence Research 16, 359387.
Wolpert D. H., Wheeler K. R. & Tumer K. 2000. Collective intelligence for control of distributed dynamical systems. EPL (Europhysics Letters) 49(6), 708.
Wooldridge M. 2001. Introduction to Multiagent Systems. John Wiley & Sons Inc.
Recommend this journal

Email your librarian or administrator to recommend adding this journal to your organisation's collection.

The Knowledge Engineering Review
  • ISSN: 0269-8889
  • EISSN: 1469-8005
  • URL: /core/journals/knowledge-engineering-review
Please enter your name
Please enter a valid email address
Who would you like to send this to? *
×

Metrics

Full text views

Total number of HTML views: 3
Total number of PDF views: 40 *
Loading metrics...

Abstract views

Total abstract views: 163 *
Loading metrics...

* Views captured on Cambridge Core between 24th August 2017 - 22nd January 2018. This data will be updated every 24 hours.