Abstract
Teacher-student reinforcement learning is a popular approach that aims to accelerate the learning of new agents with advice from trained agents. In these methods, budgets are introduces to limit the amount of advice to prevent over-advising. However, existing budget-based methods tend to use up budgets in the early training stage to help students learn initial policies fast. As a result, initial policies are some kind solidified, which is not beneficial for improving policy generalization. In this paper, to overcome advising intensively in the early training stage, we enable advising in the entire training stage in a decreasing way. Specifically, we integrate advice into reward signals and propose an advice-based extra reward method, and integrate advice into exploration strategies and propose an advice-based modified epsilon method. Experimental results show that the proposed methods can effectively improve the policy performance on general tasks, without loss of learning speed.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Amir, O., Kamar, E., Kolobov, A., Grosz, B.J.: Interactive teaching strategies for agent training. In: Proceedings of the Twenty-Fifth International Joint Conference on Artificial Intelligence, pp. 804–811. AAAI Press (2016)
Arnold, K.D.: Academic achievement - a view from the top. The Illinois valedictorian project. Academic Achievement, p. 76 (1993)
Buşoniu, L., Babuška, R., De Schutter, B.: Multi-agent reinforcement learning: an overview. In: Srinivasan, D., Jain, L.C. (eds.) Innovations in Multi-agent Systems and Applications - 1. SCI, vol. 310, pp. 183–221. Springer, Heidelberg (2010). https://doi.org/10.1007/978-3-642-14435-6_7
Clouse, J.A.: On integrating apprentice learning and reinforcement learning (1997)
Cruz, F., Magg, S., Weber, C., Wermter, S.: Improving reinforcement learning with interactive feedback and affordances. In: 4th International Conference on Development and Learning and on Epigenetic Robotics, pp. 165–170. IEEE (2014)
Cruz, F., Magg, S., Weber, C., Wermter, S.: Training agents with interactive reinforcement learning and contextual affordances. IEEE Trans. Cogn. Dev. Syst. 8(4), 271–284 (2016)
Griffith, S., Subramanian, K., Scholz, J., Isbell, C.L., Thomaz, A.L.: Policy shaping: integrating human feedback with reinforcement learning. In: Advances in Neural Information Processing Systems, pp. 2625–2633 (2013)
Ilhan, E., Gow, J., Perez-Liebana, D.: Teaching on a budget in multi-agent deep reinforcement learning. In: 2019 IEEE Conference on Games, pp. 1–8. IEEE (2019)
Matignon, L., Laurent, G.J., Le Fort-Piat, N.: Independent reinforcement learners in cooperative Markov games: a survey regarding coordination problems. Knowl. Eng. Rev. 27(1), 1–31 (2012)
OpenAI: Openai five (2018). https://blog.openai.com/openai-five/
Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. MIT Press, Cambridge (2018)
Taylor, M.E., Carboni, N., Fachantidis, A., Vlahavas, I., Torrey, L.: Reinforcement learning agents providing advice in complex video games. Connect. Sci. 26(1), 45–63 (2014)
Torrey, L., Taylor, M.: Teaching on a budget: agents advising agents in reinforcement learning. In: Proceedings of the 2013 International Conference on Autonomous Agents and Multi-agent Systems, pp. 1053–1060 (2013)
Zimmer, M., Viappiani, P., Weng, P.: Teacher-student framework: a reinforcement learning approach. In: AAMAS Workshop Autonomous Robots and Multirobot Systems (2014)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2020 Springer Nature Switzerland AG
About this paper
Cite this paper
Xudong, G., Hongda, J., Xing, Z., Dawei, F., Bo, D., Jie, X. (2020). Improving Policy Generalization for Teacher-Student Reinforcement Learning. In: Li, G., Shen, H., Yuan, Y., Wang, X., Liu, H., Zhao, X. (eds) Knowledge Science, Engineering and Management. KSEM 2020. Lecture Notes in Computer Science(), vol 12275. Springer, Cham. https://doi.org/10.1007/978-3-030-55393-7_4
Download citation
DOI: https://doi.org/10.1007/978-3-030-55393-7_4
Published:
Publisher Name: Springer, Cham
Print ISBN: 978-3-030-55392-0
Online ISBN: 978-3-030-55393-7
eBook Packages: Computer ScienceComputer Science (R0)