Non linear classifier
Identify, classify, detect novelty, cope with non linearity, report cases of uncertainty.
Learning on the chip, intrinsic deduplication, tracability of the knowledge built by the neurons.
Fixed latency regardless of the number of neurons, µseconds per pattern, <150 mWatts
Available on ASICs, SOCs, and FPGAs and can be evaluated on a variety of platforms
Intel Quark SE
- 128 neurons
- 128 bytes of memory per neuron
- Component of an SOC also combining an x86 MCU and a sensor subsystem