Q-Learning Reinforcement Learning
Q-Learning & Black-Scholes
Did Finance Oversleep a Century of Development in Physics?
Machine Learning & Investing : The Good, The Bad & The Ugly
Dr. Igor Halperin & The Investment Case For Reinforecement Learning Inverse Reinforcement Learning & The Dangers Of Deep Learning
Dr. Igor Halperin & The Investment Case For Reinforecement Learning Inverse Reinforcement Learning & The Dangers Of Deep Learning
Igor Halperin is an AI Research Associate at Fidelity Investments. His research focuses on using methods of reinforcement learning, information theory, and physics for financial problems such as portfolio optimization, dynamic risk management, and inference of sequential decision-making processes of financial agents. Igor has an extensive industrial and academic experience in statistical and financial modeling, in particular in the areas of option pricing, credit portfolio risk modeling, portfolio optimization, and operational risk modeling. Prior to joining Fidelity, Igor worked as a Research Professor of Financial Machine Learning at NYU Tandon School of Engineering.
Before that, Igor was an Executive Director of Quantitative Research at JPMorgan, and a quantitative researcher at Bloomberg LP. Igor has published numerous articles in finance and physics journals, and is a frequent speaker at financial conferences. He has co-authored the books “Machine Learning in Finance: From Theory to Practice” (Springer 2020) and “Credit Risk Frontiers” (Bloomberg LP, 2012). Igor has a Ph.D. in theoretical high energy physics from Tel Aviv University. And a M.Sc. in nuclear physics from St. Petersburg State Technical University.
Reinforcement learning (RL) is an area of machine learning concerned with how intelligent agents ought to take actions in an environment in order to maximize the notion of cumulative reward. Moreover, reinforcement learning is one of three basic machine learning paradigms, alongside supervised learning and unsupervised learning. In addition, reinforcement learning differs from supervised learning in not needing labelled input/output pairs be presented, and in not needing sub-optimal actions to be explicitly corrected. Instead the focus is on finding a balance between exploration (of uncharted territory) and exploitation (of current knowledge).
In conclusion, the environment is typically stated in the form of a Markov decision process (MDP), because many reinforcement learning algorithms for this context use dynamic programming techniques. The main difference between the classical dynamic programming methods and reinforcement learning algorithms. Lastly, is that the latter do not assume knowledge of an exact mathematical model of the MDP. And they target large MDPs where exact methods become infeasible.
