Neuromorphic AI Advantages and Recent Progress

Neuromorphic AI is developing spiking Neural Networks. Spiking neural networks can run different algorithms than neural networks. They have temporal properties and features.

It requires less data and less energy than regular neural networks. Energy can drop 100 to 1000 times.

Spiking neural networks thrives where it responds to real-world inputs in real time settings.

It separates compute and memory.

Intel Labs’ second-generation neuromorphic research chip, codenamed Loihi 2, and Lava, an open-source software framework, will drive innovation and adoption of neuromorphic computing solutions.
Enhancements include:

Up to 10x faster processing capability
Up to 60x more inter-chip bandwidth
Up to 1 million neurons with 15x greater resource density
3D Scalable with native Ethernet support
A new, open-source software framework called Lava
Fully programmable neuron models with graded spikes
Enhanced learning and adaptation capabilities

Intel now has Kapoho Point. It is a compact development board for neuromorphic computing that features eight Loihi 2 chips.

Kapoho Point can represent “up to a million neurons” and “up to a billion synapses”. It can solve optimization problems (an area of particular strength for neuromorphic computing) with up to 8 million variables. It is up to 1000× better energy efficiency than a state-of-the-art CPU solver.

4 thoughts on “Neuromorphic AI Advantages and Recent Progress”

  1. I am guessing it’s an American trillion you are referencing therefore only 6 orders of magnitude from 1 million. And you forget to reference the cycle rate which is much higher in a computer. Also one board only unlike is they can and do stack.

    They will outperform us sometime soon it is inevitable. And very Borg!

Comments are closed.