Combining Quantum LSTM Forecasts with Reinforcement Learning for FX Trading

Summary

The document describes a long-only USD/TWD trading agent that combines Quantum Long Short-Term Memory (QLSTM) forecasts of short-term trends with Quantum Asynchronous Advantage Actor-Critic (QA3C). Its state uses QLSTM features and technical indicators, while its reward is designed to support trend-following and risk control. The reported setup uses a sequence length of four and eight QA3C workers, with data spanning 2000 to 2025 split into training and test sets.

The authors report an 11.87% return over about five years and a maximum drawdown of 0.92%, and say the agent outperformed several currency ETFs. These results are presented as evidence that the hybrid approach may suit small-profit trades with tight risk limits. The document gives only a brief account of the experiment, without detailed comparisons, statistical uncertainty, or enough information to assess robustness. It also notes that the quantum methods were run through classical simulation and that the strategy was simplified, so the findings do not establish an advantage from quantum hardware or generalize automatically to other markets.

Key ideas

• The agent combines QLSTM short-term trend forecasts with QA3C reinforcement learning for USD/TWD trading.
• Its state includes forecast features and technical indicators, and its reward accounts for trend-following and risk control.
• The reported long-only results include an 11.87% return over about five years and a 0.92% maximum drawdown.
• The quantum components were simulated classically, and the authors describe the strategy as simplified.

Tags

# Quantum-Enhanced Forecasting for Deep Reinforcement Learning in Algorithmic Trading


# Quantum-Enhanced Forecasting for Deep Reinforcement Learning in Algorithmic Trading









The convergence of quantum-inspired neural networks and deep reinforcement learning offers a promising avenue for financial trading. We implemented a trading agent for USD/TWD by integrating Quantum Long Short-Term Memory (QLSTM) for short-term trend prediction with Quantum Asynchronous Advantage Actor-Critic (QA3C), a quantum-enhanced variant of the classical A3C. Trained on data from 2000-01-01 to 2025-04-30 (80\% training, 20\% testing), the long-only agent achieves 11.87\% return over around 5 years with 0.92\% max drawdown, outperforming several currency ETFs. We detail state design (QLSTM features and indicators), reward function for trend-following/risk control, and multi-core training. Results show hybrid models yield competitive FX trading performance. Implications include QLSTM's effectiveness for small-profit trades with tight risk and future enhancements. Key hyperparameters: QLSTM sequence length$=$4, QA3C workers$=$8. Limitations: classical quantum simulation and simplified strategy. \footnote{The views expressed in this article are those of the authors and do not represent the views of Wells Fargo. This article is for informational purposes only. Nothing contained in this article should be construed as investment advice. Wells Fargo makes no express or implied warranties and expressly disclaims all legal, tax, and accounting implications related to this article.

Shown in full with attribution under the source's licence. Licence: abstract CC0

This summary was written by Stratmill's research agent from the original; it is not a copy of the source.