Intel Labs launches 'Kapoho Point' neuromorphic board

Intel Labs launches ‘Kapoho Point’ neuromorphic board

Over the past five years, Intel has been iterating on its neuromorphic chips and systems, with the goal of creating devices (and software for those devices) that closely mimic the behavior of the human brain through the use of computational “neurons.” . When Loihi 2 was announced, it was only accessible through the Intel Neuromorphic Research Cloud to select members of the Intel Neuromorphic Research Community (INRC); now, the availability of Loihi 2 is expanding with the announcement of Kapoho Point, a compact development board for neuromorphic computing that includes eight Loihi 2 chips.

“We launched the Loihi neuromorphic chip five years ago,” said Mike Davies, principal engineer and director of Intel’s Neuromorphic Computing Lab. “Over the past five years, we have built a number of different form factor boards and systems using the Loihi chip, ranging from small USB-like form factor devices to large mounted enclosures. racked… with 768 Loihi chips in it.”

Loihi 2. Image courtesy of Intel.

“A year ago we announced the Loihi 2 chip, and it improves on Loihi 1 in almost every dimension,” Davies continued. “it’s faster; it’s higher bandwidth in its inter-chip communications, so we can scale and solve bigger problems much more efficiently now; it has more capacity per chip and the chip is smaller through process scaling; [and] we have a lot more programmability in the chip[.]”

Kapoho Point (which, like all of Intel’s neuromorphic releases so far, is named after Hawaii’s volcanic geography) marks a new form factor for Intel: an ultra-compact board about three inches square (Davies l compared to a credit card) with four Loihi 2 chips on its upper side and four more on its lower side, for a total of eight Loihi 2 chips at “the most minimal scale possible”.

Kapoho Point. Image courtesy of Intel.

Davies said Kapoho Point can represent “up to a million neurons” and “up to a billion synapses” – “a pretty good scale of network size just in this very compact form factor.” It can solve optimization problems (a particularly strong area for neuromorphic computing) with up to 8 million variables and, according to Intel, with up to 1000 times the power efficiency of a CPU solver at the cutting edge of technology.

“Another property you’ll see in this diagram about Kapoho Point is that it’s stackable,” Davies said. “You can just stack them through that expansion port you see on the board over there and expand to a bigger capacity. We’ve assembled stacks like [high] like four cards here that get up to 32 million neurons, which is a pretty reasonable scale, especially considering the optimization issues. Davies said Intel expects Kapoho Point to be stackable up to eight boards before other form factors become more desirable solutions.

Stacked Kapoho Point boards. Image courtesy of Intel.

Davies said Intel was excited to see where Kapoho Point would be used by the research community, pointing to potential applications in areas such as robotics and drones. Intel’s Loihi and Loihi 2 chips have been used to date in applications ranging from robotic control and gesture recognition to scene understanding and scent identification.

Kapoho Point has already delivered to select partners (it’s a “true working system,” Davies said). The Air Force Research Laboratory (AFRL) was the first research partner to receive the board, and Intel says AFRL is using it to research neural network-based learning and real-time optimization. . “Because AFRL’s missions are in the air and in space, mobile platforms have limited space, weight and power budgets,” said Qing Wu, lead scientist for processing and operation at the AFRL. “Neuromorphic computing technology gives us the best computing solution to run AI algorithms in the environment.”

Along with the Kapoho Point announcement, Intel highlighted recent additions to its Lava software framework aimed at enabling Loihi 2. The company also unveiled the latest round of its sponsored INRC projects, through which Davies said Intel is “funding university groups to help us explore the frontier”. neuromorphic algorithms and application development using the Loihi architecture. Recipients include Brown University; ETH Zürich; George Mason University; Graz University of Technology; Pennsylvania State University; Queensland University of Technology; the University of Göttingen; and the University of Waterloo. Research areas range from auditory feature detection to brain-computer interfaces.

#Intel #Labs #launches #Kapoho #Point #neuromorphic #board

Leave a Comment

Your email address will not be published. Required fields are marked *