Computational Stochastic Optimization and Learning

CASTLE Labs works to advance the development of modern analytics for solving a wide range of applications that involve decisions under uncertainty. 

See article from BBC Future on the math problem that modern life depends on.

New book!  Reinforcement Learning and Stochastic Optimization: A unified framework for sequential decisions

  • Core activities span modeling, computation and theory. Our applications span e-commerce, energy, health, and transportation.  We once worked on optimal learning in materials science.  We are starting to work on autonomous systems including drones and robots.
  • We use a unified framework that spans 15 different communities that all deal with decisions under uncertainty.  This framework that allows us to model any sequential decision problem in the presence of different sources of uncertainty.
  • Teaching – We now teach this framework to both undergraduates and graduate students across a range of departments.
  • Today, there is considerable confusion about the meaning of terms like “artificial intelligence,” and “reinforcement learning.”  Click here for our own explanation of what is AI.

Surrounding the core activities in methodology are laboratories focusing on major areas of application:

  • Optimal learning – This research addresses the challenges of collecting information, when information (observations, simulations, laboratory and field experiments) are expensive. See our new hOLMES Data Sciences website for learning in materials science.
  • Health sciences – Projects in health have included drug discovery, drug delivery, blood management, dosage decisions, personal health, and health policy.

I hope you find the material interesting, and perhaps useful. If you have any questions, please contact me.

Warren Powell