⚡ 本页包含 AI 生成的分析内容,仅供参考
本文提出ENVISION处理器,通过次字并行和动态电压精度频率可扩展技术,实现了0.26至10TOPS/W的宽能效范围,解决了CNN推理中能效与灵活性的权衡问题。
ConvNets, or Convolutional Neural Networks (CNN), are state-of-the-art classification algorithms, achieving near-human performance in visual recognition
Bert Moons, Roel Uytterhoeven, Wim Dehaene, Marian Verhelst
KU Leuven, Leuven, Belgium