The ambition of this page is to accumulate a comprehensive list of the many large and small success stories in RL and thereby dispel the myth that RL has had very few success stories since TDgammon.
Of course, we first need to define what one means by a success story. One possibility is to only consider fielded applications. That seems problematic not only because it is difficult to know what counts for a fielded application (e.g., if someone uses RL on a robot - does that count as a fielded application?) but also because fielding an application requires skills that are often irrelevant to the RL effort. So a more reasonable definition of success story might be that it involves the application of RL methods to a substantial and difficult problem domain that is of independent interest to some other community. Yes, this is vague and therefore may include applications that some of you may not want to count towards success stories. Regardless my hope is that we can collect enough such stories to convincingly squash this myth.
There is a separate page that is collecting Successes of RL. Please help by e-mailing your own list of success stories.