IBM unveiled the NorthPole chip to accelerate AI-based image recognition tasks

IBM Research

A team of scientists and engineers from IBM Research has developed a specialised NorthPole chip that can run AI-based image recognition applications 22 times faster than existing chips on the market.

Here's What We Know

As experts point out, the use of "edge systems" (edge systems), where data is physically closer to AI applications, allows them to be faster.

At NorthPole, engineers have combined the processing module and the data it uses. The chip's design is inspired by the workings of the human brain, the researchers say.

NorthPole uses a two-dimensional array of memory blocks and processors to perform tasks. The architecture allows for efficient communication between them and fast processing of information.

In testing, NorthPole outperformed other chips, including NVIDIA GPUs, by a factor of 22 on typical image recognition tasks.

However, engineers admit that the accelerator is limited to highly specialised AI operations. IBM plans to test clusters of multiple NorthPole chips to expand its capabilities.

Source: TechXplore