An Introduction to Planning with Reinforcement Learning

Founding member, and the second person (https://www.meetup.com/nyhackr/events/10251302/) to ever to present to the group, Shane Conway has returned to New York and will be telling us all about reinforcement learning.

This month's meetup is sponsored by O'Reilly and they are offering our group a 20% discount to the Strata Conference next month by visiting http://oreil.ly/UGSHW14 and using code UGNYHACKR20.

http://photos1.meetupstatic.com/photos/event/a/9/6/c/600_407683372.jpeg

About the talk:

Machine learning is often divided into three categories: supervised, unsupervised, and reinforcement learning. Reinforcement learning concerns problems with sequences of decisions (where each decision affects subsequent opportunities), in which the effects can be uncertain, and with potentially long-term goals. It has achieved immense success in various different fields, especially AI/Robotics and Operations Research, by providing a framework for learning from interactions with an environment and feedback in the form of rewards and penalties.

Shane will give a general overview of the subject of reinforcement learning, covering how to solve cases where there is uncertainty both in actions and states, as well as where the state space is very large. He will also walk through several code examples in R for how to apply this to specific problems.

Bio:

Shane is a researcher at Kepos Capital, a quantitative Global Macro firm located in New York City. His research interests have largely focused on optimal portfolio transactions, which has led to a focus on sequential decision theory and computational learning theory. He has a degree in Electrical Engineering from Columbia University. He can be reached through twitter @statalgo (https://twitter.com/statalgo).

Pizza starts at 6:30, giveaways and the talk begin at 7, then after we will go to the bar.