One of the big challenges of current electronics is the design and implementation of hardware neural networks that perform fast and energy-efficient machine learning. Spintronics is a promising catalyst for this field with the capabilities of nanosecond operation and compatibility with existing microelectronics. Considering large-scale, viable neuromorphic systems however, variability of device properties is a serious concern.
Jan Kaiser, Purdue University, and colleagues reveal in a recently-published paper from PRApplied that in situ learning of weights and biases in a Boltzmann machine can counter device-to-device variations and learn the probability distribution of meaningful operations such as a full adder. This scalable autonomously operating learning circuit using spintronics-based neurons could be especially of interest for standalone artificial-intelligence devices capable of fast and efficient learning at the edge.
Kaiser sits down with the Physical Review Journal Club to discuss this device and the results from their recent study. After a brief presentation, Kaiser will answer all participant questions in a session moderated by Matthew Daniels, NIST-Gaithersburg. Registration is free and a video recording will be sent to all registrants.
