Collaboration to Develop Neuromorphic Chips Should Accelerate Development

Friday, October 14, 2016

Neuromorphic Computing

Neuromorphic Computing

Stanford University and Korean chip maker SK Hynix have entered into a partnership agreement, with the aim being the creation of a neuromorphic chip.  The group is planning to develop the Artificial Neural Network Devices utilizing using newly discovered ferroelectric materials. The development could be a big step for the development of neuromorphic computing.

Korean chip maker SK Hynix Inc. has announced it has entered into an agreement with Stanford University to collaboratively research and develop ‘Artificial Neural Network Devices’ exploiting a ferroelectric material. This collaboration is expected to mark a milestone in the development of neuromorphic chips.

"The R&D effort will become the touchstone for future artificial intelligence era."
Lam Research Corporation and Versum Materials, Inc. have also joined to accelerate the development process.

Neuromorphic chips are semiconductors that function almost as imitations of the thinking process of the human brain. In the Big Data era, with numerous unstructured data like text, image, voice and video are combined, there is a limit in data recognition with standard serial order data processing.

The replacement of standard Von Neumann chips by neuromorphic chips is also expected to bolster the growth of neuromorphic and cognitive computing industry exponentially in the coming years.

Collaboration to Develop Neuromorphic Chips Should Accelerate Development
Dr. John Langan CTO(Chief Technology Officer) of Versum Materials Inc., Prof. Philip Wong of Stanford Univ., Prof. Yoshio Nishi of Stanford Univ., Dr. Choi Yongsoo of SK Hynix Inc., Dr. Dave Hemker CTO of Lam Research Co.

A neuromorphic chip is perceived to be the most efficient in unstructured data processing based on current research and development. The chip architecture also makes up for speed drops and helps to dramatically minimize power consumption, in another way of how the technology mimics the human brain.

Related articles
This collaboration is expected to be an important starting point for going beyond the current limitation of computing structure in data throughput and speed. Current computing systems fulfill a command in serial order in which it is entered into a logic semiconductor such as CPU (Central Processing Units) or AP (Application Processors) then sequentially delivered to a memory semiconductor such as DRAM and NAND Flash.

Neuromorphic chips, on the other hand, create an entirely new data processing system. A neuromorphic chip will be equipped with both logic operation as well as memory functions, and therefore available for multiple simultaneous logic and data processes in a manner similar to how a human brains responds to external stimulus.

“This joint R&D will make the best use of all the participants from devices, manufacturing process, equipment, materials, architecture area to accelerate the development process of the Artificial Neural Network Devices,”, said Executive Vice President Sung Joo Hong, the Head of R&D, SK Hynix.

“Plenty of research results of the ferroelectric materials have been accumulated from an academic perspective so the degree of understanding of which is deep enough to have bright prospect for the joint R&D,”, said Professor Yoshio Nishi of Stanford University. “The R&D effort will become the touchstone for future artificial intelligence era,” he added.


By  33rd SquareEmbed


Post a Comment