Deep Reinforcement Learning for Trading- ON-339Desired discipline(s): Engineering - other, Engineering, Computer science, Mathematical Sciences, Finance, Mathematics, Operations research, Statistics / Actuarial sciences
Project Length: 4 to 6 months
Preferred start date: As soon as possible.
Language requirement: English
Location(s): Toronto, ON, Canada; Canada; Canada
No. of positions: 1
Preferred institutions: McGill University, Polytechnique Montréal, University of British Columbia, University of Montreal, University of Toronto
About the company:
Our company is a quantitative finance research firm. We hire the brightest minds in the world to tackle some of the challenging problems in finance. We pair this expertise with artificial intelligence, big data, and some of the most advanced technology available to predict movements in financial markets.
Please describe the project.:
Sequential decision making under uncertainty arises in many areas such as navigation, process control, medical diagnosis and finance. A common way to solve such problems is the reinforcement learning (RL) framework, in which an agent learns to make decisions by interacting with its environment. Trading financial securities using current state-of-the-art RL algorithms is still a challenging problem and requires an explicit quantification of risk in the agent’s objective. Rigorous handling of different types of financial risks is not typical in the machine learning community, and entails many implementation and computational challenges. The objective of this project is to develop a more general deep RL framework based on our previous work that can autonomously learn to make more prudent investing decisions.
- PhD in experimental or theoretical science (life science, mathematics, physics, statistics etc.) or in a computational scientific field (big data, computer science, computational biology or chemistry, engineering. Strong M.Sc. students’ applications will be considered.
- Post PhD experience (academic or private sector research).
- Ability for analysis of complex data sets, modelling and practical implementations in simulation environments.
- Programming skills in Python, C++.
- Comprehensive knowledge and relevant experience in deep learning and reinforcement learning.
- Adaptable and rigorous, capable of working in a quickly evolving environment.
- Strong teamwork and communication skills.