🤖 AI Summary
To address the critical bottlenecks of low energy efficiency, high noise susceptibility, and volatility in cryogenic machine learning hardware for quantum computing applications—such as quantum error correction and control—this work introduces a novel cryogenic in-memory computing paradigm based on magnetic topological insulators. We pioneer the magnetic topological memristor, achieving for the first time electrically tunable chirality of helical edge states via spin–momentum-locked current. Leveraging the giant bipolar anomalous Hall effect, we enable high signal-to-noise-ratio, nonvolatile readout. Through cryogenic memristor array integration and hardware–software co-design of in-memory neural networks, our prototype achieves high accuracy on proof-of-concept classification tasks. Large-scale image recognition and quantum state preparation simulations demonstrate significantly improved accuracy and over 50% reduction in energy consumption compared to state-of-the-art memristive technologies.
📝 Abstract
Energy-efficient hardware implementation of machine learning algorithms for quantum computation requires nonvolatile and electrically-programmable devices, memristors, working at cryogenic temperatures that enable in-memory computing. Magnetic topological insulators are promising candidates due to their tunable magnetic order by electrical currents with high energy efficiency. Here, we utilize magnetic topological insulators as memristors (termed magnetic topological memristors) and introduce a chiral edge state-based cryogenic in-memory computing scheme. On the one hand, the chiral edge state can be tuned from left-handed to right-handed chirality through spin-momentum locked topological surface current injection. On the other hand, the chiral edge state exhibits giant and bipolar anomalous Hall resistance, which facilitates the electrical readout. The memristive switching and reading of the chiral edge state exhibit high energy efficiency, high stability, and low stochasticity. We achieve high accuracy in a proof-of-concept classification task using four magnetic topological memristors. Furthermore, our algorithm-level and circuit-level simulations of large-scale neural networks based on magnetic topological memristors demonstrate a software-level accuracy and lower energy consumption for image recognition and quantum state preparation compared with existing memristor technologies. Our results may inspire further topological quantum physics-based novel computing schemes.
