发表机构
National Taiwan Normal University; National Taiwan University; PecuLab LLC; Wells Fargo（国立台湾师范大学; 国立台湾大学; PecuLab有限责任公司; 富国银行）
机构由 AI 辅助整理，请以论文原文为准。AI 中文总结
本研究提出 Titans-QFWP 混合强化学习架构，结合量子快速权重编程器与 Titans 记忆机制，采用增强型 A3C² 框架，在 468 只标普 500 股的 EPC 基准下取得优异投资组合优化性能，可兼顾回撤控制与上行潜力。
AI 中文摘要
我们提出了Titans-QFWP，这是一种混合强化学习架构，将量子快速权重编程器（Quantum Fast Weight Programmer）与 Titans 风格的记忆机制（包含持久性、惊喜性和遗忘性）相结合，用于自适应投资组合优化。为解决高维市场特征问题，我们引入了增强型 A3C² 框架，该框架采用匈牙利算法对齐的 K-means 聚类和缩放对数收益奖励函数。在 468 只标普 500 成分股上，于参数数量相当（Equal-Parameter-Count，EPC）基准下，使用约 3000 个可训练参数，Titans-QFWP 取得了优异性能：中位数年化收益率（ARR）为 0.4260，卡玛比率（Calmar）为 8.5504，信息比率（IR）为 0.8427。 ablation 实验结果表明，量子门控从根本上改变了记忆组件的作用：持久性支持回撤控制，惊喜性助力收益生成，遗忘性提供额外稳定性。通过稳定这些量子表示，该模型在市场回撤期间能实现防御性配置，同时保留上行潜力。
英文摘要
We propose Titans-QFWP, a hybrid reinforcement learning architecture integrating a Quantum Fast Weight Programmer with Titans-style memory (Persistence, Surprise, and Forgetting) for adaptive portfolio optimization. To address high-dimensional market features, we introduce an enhanced A3C^2 framework with Hungarian-aligned K-means clustering and scaled log-return rewards. Evaluated on 468 S&P 500 stocks under an Equal-Parameter-Count (EPC) benchmark with approximately 3,000 trainable parameters, Titans-QFWP achieves strong performance (median ARR 0.4260, Calmar 8.5504, IR 0.8427). Ablation results reveal that quantum gating fundamentally reshapes memory component roles, with Persistence supporting drawdown control, Surprise contributing to return generation, and Forgetting providing additional stabilization. By stabilizing these quantum representations, the model enables defensive allocation during market drawdowns while preserving upside potential.
Comments4 pages, 3 figures, 4 tables, accepted for presentation at the IEEE International Conference on Quantum Computing and Engineering (QCE) 2026 QCRL Workshop
