MAPS: Multi-agent Reinforcement Learning-based Portfolio Management System
This work addresses portfolio management for investors by offering a method to adapt to changing market conditions, though it is incremental as it builds on existing multi-agent and reinforcement learning approaches.
The paper tackles the problem of generating investment strategies in stock markets by proposing MAPS, a multi-agent reinforcement learning system where each agent acts as an independent investor to create diversified portfolios, and it shows that MAPS outperforms most baselines in terms of Sharpe ratio using 12 years of US market data.
Generating an investment strategy using advanced deep learning methods in stock markets has recently been a topic of interest. Most existing deep learning methods focus on proposing an optimal model or network architecture by maximizing return. However, these models often fail to consider and adapt to the continuously changing market conditions. In this paper, we propose the Multi-Agent reinforcement learning-based Portfolio management System (MAPS). MAPS is a cooperative system in which each agent is an independent "investor" creating its own portfolio. In the training procedure, each agent is guided to act as diversely as possible while maximizing its own return with a carefully designed loss function. As a result, MAPS as a system ends up with a diversified portfolio. Experiment results with 12 years of US market data show that MAPS outperforms most of the baselines in terms of Sharpe ratio. Furthermore, our results show that adding more agents to our system would allow us to get a higher Sharpe ratio by lowering risk with a more diversified portfolio.