用可调非线性连接构建低功耗模拟神经网络，适合连续控制任务
Low-power analogue neural networks with trainable nonlinear connections for continuous control
- 将可训练非线性函数置于连接上，取代传统权重，提升计算效率
- 在机器人运动控制等连续任务中，节点和连接数比多层感知机少得多
- 适用于硬件部署，35,000个连接下可迁移，功耗可降至30微瓦
物理神经网络通过直接利用模拟器件物理特性实现低功耗机器学习，但多数架构迫使非线性器件响应充当标量权重。受柯尔莫哥洛夫-阿诺德网络启发，我们将可训练的非线性函数置于连接上，使每个物理连接成为可学习的计算单元。在可编程模拟阵列上以模拟带通滤波器实现这些函数，发现其优势依赖于任务类型，源于物理基函数的平滑性：网络能以更少节点和连接表示平滑的连续目标，如机器人运动学、连续控制和光伏最大功率点跟踪，但在分类类决策边界任务中无参数效率优势。训练后的网络可在约35,000个连接上实现硬件迁移并量化保真度；专用CMOS实现预计功耗为约30微瓦。忆阻器实现的仿真重现相同行为，表明优势来自连接上的可训练非线性，而非特定器件。
原文摘要 · Abstract (English)
Physical neural networks promise low-power machine learning by computing directly with analogue device physics, but most architectures force nonlinear device responses to act as scalar weights. Inspired by Kolmogorov-Arnold networks, we place trainable nonlinear functions on the connections, making each physical connection a learnable computational element. Realising these functions as analogue band-pass filters on field-programmable analogue arrays, we find that the benefit is task-dependent and follows from the smoothness of the physical basis: the networks represent smooth, continuously valued targets, including robotic kinematics, continuous control, and photovoltaic maximum-power-point tracking, with far fewer nodes and connections than multilayer perceptrons, but offer no parameter-efficiency advantage on classification-like decision boundaries. Trained networks transfer to hardware across approximately 35,000 connections with quantified fidelity, and a dedicated CMOS implementation is projected to operate at approximately 30 microwatts. A memristive realisation reproduces the same behaviour in simulation, indicating that the advantage comes from placing trainable nonlinearity on connections, rather than from a particular device.
Thank you to arXiv for use of its open access interoperability. PaperDance 不是 arXiv 官方产品；中文卡片由大模型生成，请以原文为准。