Combining Reinforcement Learning and Inverse Reinforcement Learning for Asset Allocation Recommendations

9 Pages Posted: 10 Jan 2022

See all articles by Igor Halperin

Igor Halperin

Fidelity Investments, Inc.

Jiayu Liu

Fidelity Investments, Inc.

Xiao Zhang

Fidelity Investments, Inc.

Xiao Zhang

affiliation not provided to SSRN

Date Written: January 6, 2022

Abstract

We suggest a simple practical method to combine the human and artificial intelligence to both learn best investment practices of fund managers, and provide recommendations to improve them.
Our approach is based on a combination of Inverse Reinforcement Learning (IRL) and RL. First, the IRL component learns the intent of fund managers as suggested by their trading history, and recovers their implied reward function. At the second step, this reward function is used by a direct RL algorithm to optimize asset allocation decisions. We show that our method is able to improve over the performance of individual fund managers.

Keywords: Active portfolio management, asset allocation, reinforcement learning, inverse reinforcement learning

JEL Classification: C44, C45, C51, C52, C54, C61, G02, G10, G11, G12

Suggested Citation

Halperin, Igor and Liu, Jiayu and Zhang, Xiao and Zhang, Xiao, Combining Reinforcement Learning and Inverse Reinforcement Learning for Asset Allocation Recommendations (January 6, 2022). Available at SSRN: https://ssrn.com/abstract=4002715 or http://dx.doi.org/10.2139/ssrn.4002715

Igor Halperin (Contact Author)

Fidelity Investments, Inc. ( email )

United States

Jiayu Liu

Fidelity Investments, Inc. ( email )

United States

Xiao Zhang

Fidelity Investments, Inc. ( email )

United States

Xiao Zhang

affiliation not provided to SSRN

Do you have a job opening that you would like to promote on SSRN?

Paper statistics

Downloads
1,129
Abstract Views
4,470
Rank
41,373
PlumX Metrics