Sign In

Communications of the ACM

ACM TechNews

Technologists Develop FatNet Algorithm

View as: Print Mobile App Share:

FatNet is based on the technique called convolutional neural networks (CNNs) to process and classify images. These specialized deep learning algorithms are known for their effectiveness in image recognition tasks.

Credit: City University of London (U.K.)

FatNet, an algorithm developed by researchers at the U.K.'s City, University of London, can convert any convolutional neural network into a specialized network that is more compatible with an optical artificial intelligence (AI) accelerator.

Specifically designed for deep learning, the algorithm integrates AI models into 4F free-space optical accelerators, which use less energy and are faster than traditional CPU/GPU-based AI.

FatNet could be used to improve medical diagnostic tools and autonomous vehicle technology, among other applications.

From City University of London (U.K.)
View Full Article


Abstracts Copyright © 2023 SmithBucklin, Washington, DC, USA


No entries found

Sign In for Full Access
» Forgot Password? » Create an ACM Web Account