TY - GEN
T1 - MetaTrader
T2 - 31st ACM International Conference on Information and Knowledge Management, CIKM 2022
AU - Niu, Hui
AU - Li, Siyuan
AU - Li, Jian
N1 - Publisher Copyright:
© 2022 ACM.
PY - 2022/10/17
Y1 - 2022/10/17
N2 - Portfolio management is a fundamental problem in finance. It involves periodic reallocations of assets to maximize the expected returns within an appropriate level of risk exposure. Deep reinforcement learning (RL) has been considered a promising approach to solving this problem owing to its strong capability in sequential decision making. However, due to the non-stationary nature of financial markets, applying RL techniques to portfolio optimization remains a challenging problem. Extracting trading knowledge from various expert strategies could be helpful for agents to accommodate the changing markets. In this paper, we propose MetaTrader, a novel two-stage RL-based approach for portfolio management, which learns to integrate diverse trading policies to adapt to various market conditions. In the first stage, MetaTrader incorporates an imitation learning objective into the reinforcement learning framework. Through imitating different expert demonstrations, MetaTrader acquires a set of trading policies with great diversity. In the second stage, MetaTrader learns a meta-policy to recognize the market conditions and decide on the most proper learned policy to follow. We evaluate the proposed approach on three real-world index datasets and compare it to state-of-the-art baselines. The empirical results demonstrate that MetaTrader significantly outperforms those baselines in balancing profits and risks. Furthermore, thorough ablation studies validate the effectiveness of the components in the proposed approach.
AB - Portfolio management is a fundamental problem in finance. It involves periodic reallocations of assets to maximize the expected returns within an appropriate level of risk exposure. Deep reinforcement learning (RL) has been considered a promising approach to solving this problem owing to its strong capability in sequential decision making. However, due to the non-stationary nature of financial markets, applying RL techniques to portfolio optimization remains a challenging problem. Extracting trading knowledge from various expert strategies could be helpful for agents to accommodate the changing markets. In this paper, we propose MetaTrader, a novel two-stage RL-based approach for portfolio management, which learns to integrate diverse trading policies to adapt to various market conditions. In the first stage, MetaTrader incorporates an imitation learning objective into the reinforcement learning framework. Through imitating different expert demonstrations, MetaTrader acquires a set of trading policies with great diversity. In the second stage, MetaTrader learns a meta-policy to recognize the market conditions and decide on the most proper learned policy to follow. We evaluate the proposed approach on three real-world index datasets and compare it to state-of-the-art baselines. The empirical results demonstrate that MetaTrader significantly outperforms those baselines in balancing profits and risks. Furthermore, thorough ablation studies validate the effectiveness of the components in the proposed approach.
KW - deep reinforcement learning
KW - imitation learning
KW - meta-policy learning
KW - portfolio management
UR - https://www.scopus.com/pages/publications/85140836989
U2 - 10.1145/3511808.3557363
DO - 10.1145/3511808.3557363
M3 - 会议稿件
AN - SCOPUS:85140836989
T3 - International Conference on Information and Knowledge Management, Proceedings
SP - 1573
EP - 1583
BT - CIKM 2022 - Proceedings of the 31st ACM International Conference on Information and Knowledge Management
PB - Association for Computing Machinery
Y2 - 17 October 2022 through 21 October 2022
ER -