BrainChip(ASX:BRN) is the jewel in the crown of Peter van der Made's life's work. Having previously worked for IBM, he has more recently been working to emulate the human brain, not through computer software like other large computer companies such as Microsoft and IBM, but through a hardware solution.
BrainChip's Spiking Neuron Adaptive Processor (S.N.A.P.) takes advantage of a network of neural like computer processors called motor neurons, combined with sensory chips and a central CPU.
Unlike software solutions, which require massive amounts of computer processor power and huge amounts of energy, BrainChip's hardware solution is claimed to run 5000 times faster than other neural computers while using less than 1/1000th of the power and avoiding complex software algorithms, a major limiting factor in Artificial Intelligence (A.I.) so far.
A combination of autonomous learning, speed and low power consumption have impressed a number of fortune 500 companies some of who helped set the companies target milestones. Brain Chip is working towards Intellectual Property (IP) licensing agreements through core product development to satisfy their requests.
The Industry sees potential for S.N.A.P. technology to deliver benefits as a stand-alone processor or as a co-processor solution, a significant number of new enquiries is providing ongoing encouragement.
The top 20 shareholders in the Company including the escrowed share holdings of the BrainChip directors and key management equals 75.65%, this tightly held share registry could help BrainChip to resist a hostile takeover.
The Tech Giants who have expressed a keen interest in A.I. Include Cisco Systems, IBM, Intel, Google, Microsoft, nVidia, Qualcomm and Samsung.
This has been a key factor in van der Made's decision to backdoor list on the ASX in order to maintain control of the company which also has maintained its expression of desire for early revenue.
Brainchip has cleared its first milestone of learning a race car game. This hurdle required the demonstration of SNAP technology to learn a basic race car game. While not the most impressive feat at first glance, it is impressive to realise that this was completed using only one chip consisting of 7 sensory and 4 motor neurons connected to one CPU. These chips have the potential to be layered, linked and scaled up to processing power not seen before on the planet, much faster and more efficient than anything we've ever seen.
"The Race Car Demonstration
In this demonstration the BrainChip team have applied its SNAP technology into a race car and demonstrates how the SNAP technology enables neurons to learn rapidly and much faster than has previously been achieved by alternative solutions. It also proves that a process can be learned and controlled using spiking neuron based learning, the same method that is used in the human brain.
BrainChip's SNAP technology can start with "blank" neurons that contain no knowledge whatsoever, and through rapid autonomous learning produces a functional set of synaptic values that express spike timing. The interface between the BrainChip neurons and a computer is also functioning and can configure the neural network for a specific function, in this case, learning to continually driving faster around the race-track without hitting the walls. "
To get an idea of how well the SNAP technology performed, the car took only 25 seconds to learn the race track in one circuit while its opponent took 15 minutes and 24 generations. Brainchip's SNAP technology is still in development, but the applications for the technology if it becomes successful such as in smart phones, robotics, and the internet of things are endless
. Bill Gate's has said "If you invent a breakthrough in artificial intelligence, so machines can learn, that is worth 10 Microsofts.
" Could this be the kind of breakthrough he was talking about?
(20min delay)