Lightweight High

Lightweight high-performance computing focuses on developing efficient algorithms and architectures for various applications, prioritizing reduced computational cost and resource consumption without sacrificing accuracy. Current research emphasizes lightweight neural networks (e.g., using ConvNeXts, MobileNets, and efficient Transformers), optimized algorithms (like variance-reduced proximal gradient methods and rational WENO schemes), and compact data representations (such as polygon maps and sparse representations). These advancements are crucial for deploying AI and machine learning models on resource-constrained devices like mobile phones, embedded systems, and UAVs, enabling real-time processing in diverse fields including medical imaging, autonomous driving, and robotics.