Startup companies are developing a technology that involves developing special chips for artificial intelligence and machine learning. Trending in the tech world, Stealth startup Cornami revealed some details of its approach to chip design to run neural networks. The chip is designed to realize the best aspects of technology first seen in 1970.
Cornami has been operating in stealth mode and this was the first time they pulled open the curtain on some of the details of how the company’s chips work. Read on to know more about this neural net chip technology.
About Cornami
Cornami built an architecture where systolic arrays can be built of any size, any shape, on demand. These systolic arrays can be dynamically re-arranged to strange new geometries that are not squares. The company aims to supply its chips to numerous markets. The list primarily includes “edge computing. Here automobiles and consumer electronics have a special need for chips that have very responsive performance and are energy-efficient in how they run neural networks.
Co-fonder and CTO, Paul Masters, described a novel way to arrange the elements of a chip to do machine learning. Here the neural network is developed as well as inference, where the neural network performs on a constant basis to provide answers. The chip developed by Cornami decreases reminiscence and cache references, and thereby “dramatically reinforce energy, latency ,and function.
Cornami won Collection B undertaking investment of $3 million in September 2016 from Affect Project Capital. The corporate has won next investment however declined to divulge how a lot.
The Neural Net Chip
According to a representative at Cornami, the chip’s technology dates back to 1970s and 1980s, called a “systolic” array. Systolic arrays are so named for the systolic functioning of the heart. It functions like blood flow, data is pumped through those computational elements. These have a plethora of computing elements, such as a multiplier-accumulator. This leads to performing the matrix multiplications that are the fundamental compute unit of neural networks. Wires connect those elements to one another, and to memory, in a grid.
Cornami chip can process an entire neural network, and will be able to replace the various combinations. These include CPUs, GPUs, FPGAs and ASICs typically used to run neural networks. It is a data center on a chip, with big implications for putting AI in edge computing such as automobiles.
The neural net chip’s inflexible association, transfers the knowledge out and in of the ones compute parts takes up a huge quantity of the chip’s effort extra effort in truth, than the compute itself inside of each and every compute part. Another characteristic of this chip is, it processes 877 frames per second in the neural network using only 30 watts. It closest competitor is Nvidia “Titan V” GPU, which processes only 8.6 frames per second at 250 watts.
SEE ALSO: “Robots For All in 5 Years” says Japan’s Tech Startup
The neural net chip is definitely a step forward in the world of technology considering the emergence of AI.
Subscribe to iTMunch for the latest news, developments and updates in the world of technology and finance.
