Hybrid learning in signalling games

Lewis–Skyrms signalling games have been studied under a variety of low-rationality learning dynamics. Reinforcement dynamics are stable but slow and prone to evolving suboptimal signalling conventions. A low-inertia trial-and-error dynamical like win-stay/lose-randomise is fast and reliable at finding perfect signalling conventions but unstable in the context of noise or agent error. Here we consider a low-rationality hybrid of reinforcement and win-stay/lose-randomise learning that exhibits the virtues of both. This hybrid dynamics is reliable, stable and exceptionally fast.

[1]  A. Roth,et al.  Learning in Extensive-Form Games: Experimental Data and Simple Dynamic Models in the Intermediate Term* , 1995 .

[2]  Geoffrey B. Sprinkle,et al.  Experimental evidence on the evolution of meaning of messages in sender-receiver games , 1998 .

[3]  Josef Hofbauer,et al.  Feasibility of communication in binary signaling games. , 2008, Journal of theoretical biology.

[4]  Jeffrey A. Barrett,et al.  Dynamic Partitioning and the Conventionality of Kinds* , 2007, Philosophy of Science.

[5]  Stanislav Volkov,et al.  Learning to signal: Analysis of a micro-level reinforcement model , 2009 .

[6]  Giosuè Baggio,et al.  Role Asymmetry and Code Transmission in Signaling Games: An Experimental and Computational Investigation , 2015, Cogn. Sci..

[7]  A. Alexandrova.  The British Journal for the Philosophy of Science , 1965, Nature.

[8]  Simon M. Huttegger,et al.  Some dynamics of signaling games , 2014, Proceedings of the National Academy of Sciences.

[9]  Kevin J. S. Zollman,et al.  Probe and Adjust in Information Transfer Games , 2014 .

[10]  Jeffrey Barrett,et al.  The role of forgetting in the evolution and learning of language , 2009, J. Exp. Theor. Artif. Intell..

[11]  A. Roth,et al.  Predicting How People Play Games: Reinforcement Learning in Experimental Games with Unique, Mixed Strategy Equilibria , 1998 .

[12]  Brian Skyrms,et al.  Self-assembling Games , 2017, The British Journal for the Philosophy of Science.

[13]  Brian Skyrms,et al.  Signals: Evolution, Learning, and Information , 2010 .

[14]  Territoire Urbain,et al.  Convention , 1955, Hidden Nature.

[15]  Mohamed A. Khamis,et al.  Enhanced multiagent multi-objective reinforcement learning for urban traffic light control , 2012, 2012 11th International Conference on Machine Learning and Applications.

[16]  Jeffrey A. Barrett,et al.  Numerical Simulations of the Lewis Signaling Game: Learning Strategies, Pooling Equilibria, and the Evolution of Grammar , 2006 .

[17]  Simon M. Huttegger.  Evolutionary Explanations of Indicatives and Imperatives , 2007 .

[18]  R. Herrnstein.  On the law of effect. , 1970, Journal of the experimental analysis of behavior.

[19]  Simon M. Huttegger,et al.  David Lewis in the lab: experimental results on the emergence of meaning , 2014, Synthese.

[20]  Kevin J. S. Zollman,et al.  Evolutionary dynamics of Lewis signaling games: signaling systems vs. partial pooling , 2007, Synthese.

[21]  B. Skyrms.  Learning to Signal with Two Kinds of Trial and Error , 2011 .