![[H]ard|Forum](/styles/hardforum/xenforo/logo_dark.png){kind=logo}
AlphaAtlas
[H]ard|Gawd
- Joined
- Mar 3, 2018
- Messages
- 1,713
Brainchip announced the "Akida" neuromorphic system on a chip today. The chip is designed for neural network acceleration, and is built on a traditional CMOS process. But, unlike competing designs that use convolution neural networks, Akida uses a spiking neural network architecture, which is supposed to bring massive power efficiency gains. While Brainchip claims to be the first company to market with this technology, Intel sampled their own neuromorphic design last year.
Each Akida NSoC has effectively 1.2 million neurons and 10 billion synapses, representing 100 times better efficiency than neuromorphic test chips from Intel and IBM. Comparisons to leading CNN accelerator devices show similar performance gains of an order of magnitude better images/second/watt running industry standard benchmarks such as CIFAR-10 with comparable accuracy. "Spiking neural networks are considered the third generation of neural networks," said Peter van der Made, Founder and CTO of BrainChip. "The Akida NSoC is the culmination of decades of research to determine the optimum neuron model and innovative training methodologies."
Each Akida NSoC has effectively 1.2 million neurons and 10 billion synapses, representing 100 times better efficiency than neuromorphic test chips from Intel and IBM. Comparisons to leading CNN accelerator devices show similar performance gains of an order of magnitude better images/second/watt running industry standard benchmarks such as CIFAR-10 with comparable accuracy. "Spiking neural networks are considered the third generation of neural networks," said Peter van der Made, Founder and CTO of BrainChip. "The Akida NSoC is the culmination of decades of research to determine the optimum neuron model and innovative training methodologies."
