Benchmarking Robustness of Deep Reinforcement Learning approaches to Online Portfolio Management - Laboratoire Interdisciplinaire des Sciences du Numérique Access content directly
Conference Papers Year : 2023

Benchmarking Robustness of Deep Reinforcement Learning approaches to Online Portfolio Management

Abstract

Deep Reinforcement Learning approaches to Online Portfolio Selection have grown in popularity in recent years. The sensitive nature of training Reinforcement Learning agents implies a need for extensive efforts in market representation, behavior objectives, and training processes, which have often been lacking in previous works. We propose a training and evaluation process to assess the performance of classical DRL algorithms for portfolio management. We found that most Deep Reinforcement Learning algorithms were not robust, with strategies generalizing poorly and degrading quickly during backtesting.

Dates and versions

hal-04473989 , version 1 (22-02-2024)

Identifiers

Cite

Marc Velay, Bich-Liên Doan, Arpad Rimmel, Fabrice Popineau, Fabrice Daniel. Benchmarking Robustness of Deep Reinforcement Learning approaches to Online Portfolio Management. 2023 International Conference on Innovations in Intelligent Systems and Applications (INISTA), Sep 2023, Hammamet, Tunisia. pp.1-6, ⟨10.1109/INISTA59065.2023.10310402⟩. ⟨hal-04473989⟩
24 View
0 Download

Altmetric

Share

Gmail Facebook X LinkedIn More