Federated reinforcement learning for portfolio management

Publication Type

Book

Publication Date

7-2022

Abstract

Financial portfolio management involves the constant redistribution of wealth over a set of financial assets and can, by its sequential nature, be modelled using reinforcement learning (RL). Federated learning allows traders to jointly train models without revealing their private data. We show on S&P500 market data how personalized, robust federated reinforcement learning using Fed+ produces trading policies that offer higher annual returns and Sharpe ratios than other methods.

Discipline

Numerical Analysis and Scientific Computing

Research Areas

Intelligent Systems and Optimization

First Page

467

Last Page

482

ISBN

9783030968960

Identifier

10.1007/978-3-030-96896-0_21

Publisher

Springer

City or Country

Cham

Additional URL

https://doi.org/10.1007/978-3-030-96896-0_21

This document is currently not available here.

Share

COinS