Skyrmion Lattices as Context-Aware Artificial Neurons

Magnetic skyrmions offer a rich physical space for hardware-aware computing, including making use of skyrmion core oscillations and skyrmion-skyrmion interactions. The behavior of arrays of skyrmions can be manipulated by multiple inputs, such as currents and electromagnetic waves. Charge current reconfigures the skyrmion lattice, thereby modulating the neuron's state, its dynamics, and its transfer function. We will show in simulation that the skyrmion lattice can therefore mimic essential advanced functionalities of the brain for adaptive intelligence, namely context-awareness, cross-frequency coupling, and feature binding. We use the skyrmion neuron to construct an adaptive artificial neural network (ANN) and show that it can perform context-aware diagnosis of breast cancer, with higher accuracy while learning faster from smaller amounts of data and using a more compact and energy-efficient network than the state-of-the-art non-adaptive ANNs used for cancer diagnosis. This work describes how hardware-based adaptive neurons can mitigate several critical challenges facing contemporary ANNs. Modern ANNs require large amounts of training data, energy and chip area and are highly task-specific. Conversely, hardware-based ANNs built with adaptive neurons show faster learning from smaller datasets, compact architectures, energy-efficiency, fault-tolerance, and can lead to the realization of general artificial intelligence.