A Deep Risk-Sensitive Reinforcement Learning Framework for Portfolio Management

In Finance, the use of Automated Trading Systems (ATS) on markets is growing every year and the trades generated by an algorithm now account for the majority of orders that arrive at stock exchanges. Historically, these systems were based on advanced statistical methods and signal processing able to extract trading signals from financial data. However, the recent successes of Machine Learning have attracted the interest of the financial community. The Reinforcement Learning method is a subcategory of machine learning and has been broadly applied by investors and researchers in building strategic asset allocation (Sequential) Decision Making systems. In this work, we explore performances of Risk-Sensitive Reinforcement Learning agents on the sequential and volatile data of financial markets by using concepts from Operations Research, Machine Learning and Finance to make the learning process of the agent immunized from adversarial impact of uncertainty.

Faculty Supervisor:

Jonathan Patrick;Jonathan Yumeng Li

Student:

Mohammad Amin Masoudi

Partner:

Brane Capital

Discipline:

Other

Sector:

Finance, insurance and business

University:

University of Ottawa