# Benchmarking Robustness of Deep Reinforcement Learning approaches to Online Portfolio Management

arXiv: https://arxiv.org/abs/2306.10950 · HTML: https://mlsanity.com/paper/2306.10950/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 10 of its 17 references that are in the map.
- First posted on arXiv 19 Jun 2023.

## Builds on (most cited first, 10 of 10)

- [Continuous control with deep reinforcement learning](https://mlsanity.com/paper/1509.02971/) (2015)
- [Deep Reinforcement Learning that Matters](https://mlsanity.com/paper/1709.06560/) (2017)
- [A Deep Reinforcement Learning Framework for the Financial Portfolio Management Problem](https://mlsanity.com/paper/1706.10059/) (2017)
- [Reinforcement-Learning based Portfolio Management with Augmented Asset Movement Prediction States](https://mlsanity.com/paper/2002.05780/) (2020)
- [Measuring the Reliability of Reinforcement Learning Algorithms](https://mlsanity.com/paper/1912.05663/) (2019)
- [Deep Reinforcement Learning for Autonomous Driving](https://mlsanity.com/paper/1811.11329/) (2018)
- [Cost-Sensitive Portfolio Selection via Deep Reinforcement Learning](https://mlsanity.com/paper/2003.03051/) (2020)
- [Bridging the gap between Markowitz planning and deep reinforcement learning](https://mlsanity.com/paper/2010.09108/) (2020)
- [Asset Allocation: From Markowitz to Deep Reinforcement Learning](https://mlsanity.com/paper/2208.07158/) (2022)
- [High-Dimensional Stock Portfolio Trading with Deep Reinforcement Learning](https://mlsanity.com/paper/2112.04755/) (2021)

## Built on by (most cited first, 1 of 1)

- [Markowitz Meets Bellman: Knowledge-distilled Reinforcement Learning for Portfolio Management](https://mlsanity.com/paper/2405.05449/) (2024)

## Abstract (arXiv)

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.