Recently, there are increasing efforts on advancing optical neural networks (ONNs), which bring significant advantages for machine learning (ML) in terms of power efficiency, parallelism, and computational speed. With the considerable benefits in computation speed and energy efficiency, there are significant interests in leveraging ONNs into medical sensing, security screening, drug detection, and autonomous driving. However, due to the challenge of implementing reconfigurability, deploying multi-task learning (MTL) algorithms on ONNs requires re-building and duplicating the physical diffractive systems, which significantly degrades the energy and cost efficiency in practical application scenarios. This work presents a novel ONNs architecture, namely, \textit{RubikONNs}, which utilizes the physical properties of optical systems to encode multiple feed-forward functions by physically rotating the hardware similarly to rotating a \textit{Rubik's Cube}. To optimize MTL performance on RubikONNs, two domain-specific physics-aware training algorithms \textit{RotAgg} and \textit{RotSeq} are proposed. Our experimental results demonstrate more than 4$\times$ improvements in energy and cost efficiency with marginal accuracy degradation compared to the state-of-the-art approaches.
翻译:RubikONNs:利用物理感知旋转架构进行多任务学习的光学神经网络
近来,越来越多的工作致力于推进光学神经网络(ONNs)的发展,ONNs在计算效率、并行性和计算速度等方面为机器学习(ML)带来了重大优势。随着计算速度和能量效率的显著提高,人们开始将ONNs应用于医疗感知、安全检测、药物检测和自动驾驶等领域。然而,由于复杂的重构问题,将多任务学习(MTL)算法部署到ONNs上需要重新构建和复制物理衍射系统,这在实际应用场景中会显著降低能量和成本效率。本文提出了一种新的ONNs架构,即RubikONNs,它利用光学系统的物理特性,物理旋转硬件来实现多个前馈函数的编码,类似于旋转光学玩具魔方。为了优化RubikONNs的MTL性能,提出了两种面向特定领域的物理感知训练算法RotAgg和RotSeq。实验结果表明,与最先进的方法相比,RubikONNs能够使能量和成本效率提高4倍以上,准确度略有下降。