互补矩阵门控QKAN快速权重编程器用于量子动力学预测
原标题：Complementary Matrix-Gated QKAN Fast-Weight Programmers for Quantum Dynamics Forecasting
AI 摘要
本文提出互补矩阵门控（CMG）机制，用于量子启发的KAN快速权重编程器，实现坐标级记忆控制，同时保持有界凸更新和仿射前缀扫描结构。在七个单步预测基准和五个序列长度上，CMG对基于QKAN的架构带来最一致的改进；在Jaynes-Cummings和transmon-resonator动力学的多步预测中，相比标量门控基线，均方误差至少降低91.2%。该工作为高效序列建模、Kolmogorov-Arnold网络和量子动力学预测之间建立了联系。
正文节选
Complementary Matrix-Gated QKAN Fast-Weight Programmers for Quantum Dynamics Forecasting Abstract Sequence models must decide what to write into memory and what to retain. In quantum and quantum-inspired sequence learning, nonlinear recurrent updates often require repeated circuit evaluations and sequential backpropagation through time, making long contexts costly. Gated fast-weight programmers (FWPs) based on quantum-inspired Kolmogorov-Arnold networks (QKANs) alleviate this bottleneck by stori