Breaking News
Trump: 'If We Didn't Have All of the Bullsh*t,' We'd Be Much Further Along
Prosecutors drop groping case against Kevin Spacey
Things Haven't Always Been This Way
Making the Worst Case Appear the Better
Top Tech News
Review: LaserCube, the world's smallest – and first – battery-powered RGB laser projector
Norshire Mini 8-inch portable air-compressor for bike, car and camping
India Lauching to the Moon in Three Days
Brain implant restores partial vision to blind people
Kanye West reportedly tried to sell SF investors on 'Star Wars'-inspired housing for homeles
Amazon Lays Out Satellite Constellation Service Goals, Deployment and Deorbit Plans to FCC
World's First 3D Printed Hearts And Functional Beating Hearts Grown From Stem Cells
Tips for protecting your food supply when the grid goes down
Baseball history made: Inside the debut of robot umpires
Ridgecrest earthquakes caused damage to Earth's crust seen from satellite
IBM Investing $2 Billion in an AI Center and Targets 1000 Times Improvement by 2029
IBM is partnering with State University of New York to develop an AI Hardware Center at SUNY Polytechnic Institute in Albany. New York will also provide a subsidy of $300 million.
The IBM Research AI Hardware Center will enable IBM and their partner ecosystem to achieve 1,000x AI performance efficiency improvement over the next decade. They will overcome current machine-learning limitations by using approximate computing with Digital AI Cores and in-memory computing with Analog AI Cores.
Approximate Computing with Digital AI Cores
The best hardware platforms for training deep neural networks (DNNs) has just moved from traditional single precision (32-bit) computations towards 16-bit precision. This is more energy efficient and uses less memory. IBM researchers have successfully trained DNNs using 8-bit floating point numbers (FP8) while fully maintaining the accuracy of deep learning models and datasets.