network connectivity patterns tagged posts

Early Bird uses 10 times Less Energy to Train Deep Neural Networks

Rice University’s Early Bird method for training deep neural networks finds key connectivity patterns early in training, reducing the computations and carbon footprint for the increasingly popular form of artificial intelligence known as deep learning. (Graphic courtesy of Y. Lin/Rice University)

Novel training method could shrink carbon footprint for greener deep learning. Engineers have found a way to train deep neural networks for a fraction of the energy required today. Their Early Bird method finds key network connectivity patterns early in training, reducing the computations and carbon footprint for training deep learning.

Early Bird is an energy-efficient method for training deep neural networks (DNNs), the form of artificial intelligence (AI) behind self-driving cars, inte...

Read More