Intel to ship first neural network processing silicon in two months

Intel says it will ship its first silicon for neural network processing before the end of this year.

Intel’s Nervana NNP

This will be the Nervana neural network processor (NNP) and Intel’s target is the growing interest in artificial intelligence (AI) computing in areas such as medical diagnosis and autonomous vehicles.

Behind the development of a range of scalability neural network processors are what Intel calls neuromorphic and quantum computing technologies.

Brian Krzanich, Intel CEO, speaking this week at the WSJDLive global technology conference:

“Neuromorphic chips are inspired by the human brain, which will help computers make decisions based on patterns and associations.”

Intel says it has developed a self-learning neuromorphic test chip, which uses data to learn and make inferences, gets smarter over time, and does not need to be trained in the traditional way.

The self-learning chips could be used to perform complex cognitive tasks, such as interpreting critical cardiac rhythms, detecting anomalies to prevent cyberhacking and composing music.

Quantum computers are defined by the use of large number of qubits (quantum bits), rather than traditional binary bits, to perform calculations in parallel.

“This will enable quantum computers to tackle problems conventional computers can’t handle, such as simulating nature to advance research in chemistry, materials science and molecular modelling – creating a room temperature superconductor or discovering new drugs,” said Krzanich.

Intel has developed a 17-qubit superconducting test chip for its quantum research partner in the Netherlands. There is a plan to deliver a 49-qubit chip by the end of this year.