
Abstract: This talk will outline applications of reinforcement learning (RL) and inverse reinforcement learning (IRL) to classical problems of quantitative finance such as portfolio optimization, wealth management and option pricing. In addition to discussing RL and IRL as computational tools, I also outline their use for theoretical research into the dynamics of financial markets. In particular, I discuss how an IRL-based view of market dynamics produces new financial models with non-linear dynamics, which are able to capture market crashes and corporate defaults.
Bio: Quant/researcher with extensive experience in building cutting-edge statistical and predictive models and advanced machine learning algorithms to solve practical problems in finance and consumer analytics, especially within credit, portfolio risk modeling, capital forecast models, and optimal control models.

Igor Halperin, PhD
Title
AI Asset Management | Fidelity Investments
