M2RU: Memristive Minion Recurrent Unit for On-Chip Continual Learning at the Edge

This article introduces a novel hardware-aware recurrent unit, M2RU, designed for continual learning on edge devices. The use of memristors suggests a focus on energy efficiency and compact implementation. The research likely explores the challenges of continual learning in resource-constrained environments, such as catastrophic forgetting and efficient adaptation to new data streams. The 'on-chip' aspect implies a focus on integrating the learning process directly onto the hardware, potentially for faster inference and reduced latency.

• •Focus on continual learning at the edge.
• •Utilizes memristors for energy-efficient hardware implementation.
• •Aims for on-chip integration for faster inference.