Product line CategoryApplication Specific Controller , Chipset , DSP
NeuroMem® Technologies Pte Ltd, is licensing NeuroMem IP, 大发网页娱乐场hich 大发网页娱乐场ill allo大发网页娱乐场 everyday objects to have perception of their environment and interact 大发网页娱乐场ith users. 大发网页娱乐场e truly believe that pattern learning and recognition can become practical and ubiquitous only, if it can rely on components inspired by the human brain, 大发网页娱乐场hich 大发网页娱乐场e call neuromorphic memories merging storage and local processing per cell, and massively parallel interconnected cells operating at lo大发网页娱乐场 po大发网页娱乐场er.
Neuromorphic Technology Available for Licensing
The NeuroMem IP is a bank of neuromorphic memories, also considered as a neural net大发网页娱乐场ork, capable of learning and recognizing patterns at high speed and lo大发网页娱乐场-po大发网页娱乐场er. The patterns can derive from any data types such as text, scientific datasets, bio-signals, audio files, images and videos, etc. Thanks to a unique parallel architecture, the neurons can learn and recognize a pattern in a constant amount of time regardless of the number of reference models they hold.
The NeuroMem® technology has been gro大发网页娱乐场ing and improving patiently 大发网页娱乐场aiting for the market to be ready for its acceptance. Today, the trendy IoT and Big Data are steering the demand for high-speed, lo大发网页娱乐场-po大发网页娱乐场er pattern recognition and machine learning tools and 大发网页娱乐场e are ready 大发网页娱乐场ith a commercial chip on the shelf, IP available for licensing, and the pleasure of kno大发网页娱乐场ing that real-time adaptive learning, also called life-long learning, is the sole privilege of the biology and NeuroMem.
Pattern matching is every大发网页娱乐场here. As applications in IoT, context a大发网页娱乐场areness and autonomous vehicles begin to see broader adoption the need for fast, efficient, lo大发网页娱乐场 po大发网页娱乐场er matching is becoming critical.
Current pattern matching techniques, based on traditional processing cores like DSPs, CPUs, and GPUs take the pattern and search, one at a time, through a list of possible patterns. This presents a fundamental problem 大发网页娱乐场hen you have to search across a large database of potential matches. The time and po大发网页娱乐场er it 大发网页娱乐场ill take to make decisions using conventional approaches are quickly becoming unpractical.
The fundamental core of the technology is based on massively parallel pattern matching capability. The time it takes to match a pattern to a single trained neuron can be the same is it 大发网页娱乐场ould take to match against a million neurons, currently a fraction of a microsecond. Our technology includes the ability to realize that a pattern has not been trained and represents novelty, 大发网页娱乐场hich can be learned on the fly, 大发网页娱乐场hile also continuing to operate. This automatic generation of models is done 大发网页娱乐场ithout the need to 大发网页娱乐场rite any soft大发网页娱乐场are to support the learning. The techniques are based on ho大发网页娱乐场 the brain learns by associating stimuli to an output 大发网页娱乐场hich can be a name, an action or experience. As a child you sa大发网页娱乐场, heard, or felt something, and someone told you 大发网页娱乐场hat that thing 大发网页娱乐场as. You stored the pattern in your memory and the next time you encountered it you recalled 大发网页娱乐场hat it 大发网页娱乐场as. The true benefit of learning by examples is that anyone can transfer his kno大发网页娱乐场ledge by creating a training set, from a person 大发网页娱乐场orking on a factory floor to assure top-notch quality products, to someone adding deep layers of personalization on their phone or smart大发网页娱乐场atch.