In-memory Photonic Tensor Core Accelerator for Neural Networks-based Applications
2023 IEEE PHOTONICS SOCIETY SUMMER TOPICALS MEETING SERIES, SUM(2023)
摘要
The growth of AI applications has pushed the need for new energy-efficient high-performance accelerators for computing Neural Networks at high speed. Here, we present our Photonic Tensor Core accelerator, which leverages photonics to compute MAC operation at high speed, and Phase Change Materials for reducing energy consumption. The resulting PTC is the key component for the next generation of AI systems.
更多查看译文
关键词
Silicon Photonics,Accelerator,Phase Change Materials
AI 理解论文
溯源树
样例
生成溯源树,研究论文发展脉络
Chat Paper
正在生成论文摘要