Particle.news
Download on the App Store
3 ARTICLES
·
6mo ago

MIT Researchers Develop Photonic Processor for Ultrafast, Energy-Efficient AI Computing

The breakthrough chip performs all key deep neural network computations optically, achieving nanosecond speeds and high accuracy.

Researchers demonstrated a fully integrated photonic processor that can perform all key computations of a deep neural network optically on the chip, which could enable faster and more energy-efficient deep learning for computationally demanding applications like lidar or high-speed telecommunications.
Image

Overview

  • The photonic processor processes machine-learning computations using light, offering significant speed and energy efficiency advantages over traditional electronic hardware.
  • Researchers designed a fully integrated optical neural network capable of performing both linear and nonlinear computations on the chip itself.
  • The chip achieved over 92% inference accuracy and completed key computations in less than half a nanosecond, comparable to conventional hardware performance.
  • Fabricated using commercial CMOS foundry processes, the chip has potential for scalable manufacturing and integration with existing electronic systems.
  • Future work aims to scale the device, integrate it with real-world applications such as telecommunications and navigation, and explore algorithms leveraging optical processing benefits.