BrainChip develops accelerated artificial intelligence and machine learning software and hardware. We have commercialized spiking neural networks, a type of neuromorphic computing which simulates the functionality of the human brain.
BrainChip’s technology uses a type of neuromorphic computing called spiking neural networks (SNNs). It has many attractive characteristics, including the ability to be trained instantaneously (“one-shot learning”), high accuracy and low compute overhead. This is an important feature in the world outside of the internet, where massive datasets are not available. For instance, a police department looking for a suspect in live video streams does not have thousands of images of that suspect, nor does it have weeks to train a traditional convolutional neural network system.
A Leader in Neuromorphic Computing
Neuromorphic computing is a branch of artificial intelligence (AI) that simulates the functionality of the human neuron. At BrainChip, we have developed a revolutionary spiking neural network (SNN) technology, a type of neuromorphic computing that learns autonomously, evolves and associates information just like the human brain.
BrainChip’s technology has been designed as a one-shot learning system. It recognizes patterns in milliseconds without having to be pre-programmed. It achieves this by learning from information, and then later recognizing what it has learned.
BrainChip’s first use of this technology is BrainChip Studio, which aids law enforcement and intelligence organizations to rapidly search vast amounts of video footage and identify patterns or faces.
Our technology learns from experience, autonomously, just like a human. It does not need to be trained with millions of samples like Deep Learning, it learns a pattern instantaneously. Deep Learning networks are power-hungry and require large GPU-Server clusters and weeks of training. BrainChip’s technology learns by itself, without large datasets, and finds patterns that humans may not be aware of. This rapid learning capability opens up new possibilities to find images in video, patterns in large datasets, and hundreds of other applications where deep learning cannot be used.
Because SNNs can be implemented using regular logic functions, they are inherently high-performance and low-power. BrainChip is in the unique positon of commercializing SNNs that offer instantaneous training, low computational overhead