Autonomous Learning of Attractors for Neuromorphic Computing

This article presents a neuromorphic computing primitive using networks of coupled Wien bridge oscillators with tunable resistive couplings. The phase relationships between oscillators encode patterns, and a local Hebbian learning rule adapts the couplings, allowing learning and recall to emerge from the same analog dynamics.

Details

The researchers developed a neuromorphic computing system using networks of coupled Wien bridge oscillators and tunable resistive couplings. The phase relationships between the oscillators encode patterns, and a local Hebbian learning rule continuously adapts the couplings, allowing learning and recall to emerge from the same ongoing analog dynamics rather than separate training and inference phases. Using a Kuramoto-style phase model, they show that the learned phase patterns form attractor states, which is validated in simulation and hardware. They also realize a 2-4-2 architecture with a hidden layer of oscillators, whose bipartite visible-hidden coupling allows multiple internal configurations to produce the same visible phase states. When inputs are switched, transient spikes in energy followed by relaxation indicate how the network can reduce surprise by reshaping its energy landscape. These results support coupled oscillator circuits as a promising hardware platform for energy-based neuromorphic computing with autonomous, continuous learning.