Self-Organized Learning in Oscillatory Neural Networks with Memristive Signed Couplings
2026-07-01 • Neural and Evolutionary Computing
Neural and Evolutionary ComputingMachine Learning
AI summaryⓘ
The authors explore how networks of oscillators (neurons that work in rhythmic patterns) can be used for brain-like computing. They designed a system using memristors that can learn on its own and clean up noisy signals, like recognizing patterns despite errors. A key advance is showing how to include both positive and negative connections (weights) in these networks, which was difficult before, allowing more complex behaviors such as stable patterns where oscillators are out of sync. Their simulations and theory show that these positive and negative weight connections are crucial for sustaining certain memory patterns over time.
Oscillatory Neural NetworksMemristorsNeuromorphic ComputingInhibitory CouplingsAuto-associative MemoryHopfield ModelIsing ModelSigned WeightsPhase-coded MemoriesAnti-phase Attractors
Authors
Riley Acker, Aman Desai, Garrett Kenyon, Frank Barrows
Abstract
Oscillatory neural networks (ONNs) have emerged as a promising neuromorphic architecture, leveraging coupled dynamical systems to perform computation and represent information through phase relationships. Their interactions can be designed to support intrinsic energy-minimizing dynamics, enabling tasks such as associative memory and optimization, and positioning them as a candidate architecture for continuous learning and inference. We present a neuromorphic primitive implemented using memristive edges with inhibitory couplings as a potential design for autonomous learning, and provide circuit simulation validation that the system is capable of denoising noisy inputs on an auto-associative task. While numerical Hopfield/Ising models routinely assume signed weights, neuromorphic implementations of ONNs often fail to realize negative weights due to device and circuit constraints. A practically implementable route to inhibitory (negative) weights is particularly valuable: it expands the class of attractor structures accessible to oscillator networks beyond purely synchronous couplings, and supports phase-coded memories where anti-phase constraints are not merely transiently enforced during training but can persist autonomously after release. We provide circuit simulations and theoretical analyses demonstrating that signed effective weights are necessary for anti-phase attractors to persist autonomously.