Penn Engineers have developed the first programmable chip that can train nonlinear neural networks using light—a breakthrough ...

Penn Engineers have developed the first programmable chip that can train nonlinear neural networks using light — a breakthrough that could dramatically speed up AI training, reduce energy use and even ...

Examples like Intel's Loihi chips tend to get competitive performance out of far lower clock speeds and energy use, but they ...

The 'Ironwood' chip marks a major shift in focus for Google, as performance and efficiency are taking a backseat.

The U.S. on Monday launched an investigation into chip imports that could pave the way for tariffs on semiconductor devices and chipmaking equipment.

And for those wary of the idea of using live cells, BBB’s current chip model doesn ... into a biological network, let neurons process them, and then decode the neural response to improve ...

Broadcom Inc. today debuted two new chips, the Sian3 and Sian2M, that data center operators can use to power their optical networks. Optical networks are often used to link together the servers in ...

A deep learning framework transforms decades of solar data into a unified, high-resolution view—adjusting instruments, overcoming limitations, and helping us better understand our star. A ...