Win-Stay, Lose-Switch

From Wikipedia, the free encyclopedia

In psychology, game theory, statistics, and machine learning, Win-Stay, Lose-Switch (also Win-Stay, Lose-Shift) is a learning strategy used to model learning in decision situations. It was first invented as an improvement over randomization in bandit problems.[1] It was later applied to the prisoner's dilemma in order to model the evolution of altruism.[2]

The learning rule bases its decision only on the outcome of the previous play. Outcomes are divided into successes (wins) and failures (loses). If the play on the previous round resulted in a success, then the agent plays the same strategy on the next round. Alternatively, if the play resulted in a failure the agent switches to another action.

[edit] References

  1. ^ Robbins, H. (1952). Some aspects of the sequential design of experiments. Bulletin of the American Mathematical Society 58, 527–535.
  2. ^ Nowak, M. and K. Sigmund (1993, July 1). A strategy of win-stay, lose-shift that outperforms tit-for-tat in the Prisoner’s Dilemma game. Nature 364, 56–58.
 This game theory-related article is a stub. You can help Wikipedia by expanding it.