Trainable lifelong AI
The NeuroMem neurons are both a learning and inference engine and they perform both tasks with fixed latencies independent of the number of neurons already in use. You can learn a new example at anytime, and this example can describe a new category if you wish. It will be taken into account immediately and contribute to the next recognition. If by mistake, you (or an unsupervised algorithm) teach an example contradicting prior learning, the neurons in conflict will autonomously flag themselves as such and correct their influence field. Neurons can also be pre-trained and this means that they are loaded with a knowledge file previously built by a NeuroMem network.
The NeuroMem neurons are responsible because they are not ashamed to output when they do not know or are unsure about a classification. In applications with cost of the mistake, such behavior is much preferred than probabilities or the famous Top3 or Top5 criteria used in Deep Learning benchmarks. Acknowledgment of uncertainties enriches the decision process by suggesting more training of the neural network or the recourse to a second opinion such as another network trained with a different feature. For the best robustess, the final decision may require minimum consensus between neurons of a same or multiple networks.
The NeuroMem neurons are not a black box. When they learn a new pattern, the decision is actually taken collectively between all the neurons already committed. The next available neuron in the chain will store the new pattern in its own memory and become committed. Consequently, the content of the neurons not only represent a knowledge that they have built together, but also the log of all the patterns that have been retained as novelty at the time they were taught.
Want to learn more?
The neurons have numerous use models enabled by their collective behavior and ability to build decision spaces incrementally, fast and with guaranteed convergence (provided that they have good teachers).
Comprehension and Analytics on your live data streams and stored datasets
The NeuroMem technology allows the design of AI engine with real-time lifelong learning capabilities and a range of applications from cognitive sensors to cognitive storage.
NeuroMem-powered Smart Sensors
Power your sensor hubs to learn and make decisions locally. Whether on an industrial machinery, under the hood of a vehicle, or around a wrist, the ability to learn and classify patterns coming directly from a sensor is game-changing.
NeuroMem -powered Smart Storage
Turn your storage devices into local, secure and configurable engines to analyze your data. Whether it is text, audio or image data, you can comprehend its content as a single item or as part of a larger dataset.