4K-Memristor Analog-Grade Passive Crossbar Circuit
This work addresses the problem of enabling efficient, on-chip synaptic storage for neuromorphic hardware, representing a significant but incremental improvement in analog-grade crossbar technology.
The researchers tackled the challenge of high device variations in passive memristor crossbars for neuromorphic computing by developing a 64x64 crossbar circuit with ~99% yield and ~26% switching voltage variations, enabling 4K-pixel gray-scale pattern programming with under 4% tuning error and MNIST classification with ~1% weight import error.
The superior density of passive analog-grade memristive crossbars may enable storing large synaptic weight matrices directly on specialized neuromorphic chips, thus avoiding costly off-chip communication. To ensure efficient use of such crossbars in neuromorphic computing circuits, variations of current-voltage characteristics of crosspoint devices must be substantially lower than those of memory cells with select transistors. Apparently, this requirement explains why there were so few demonstrations of neuromorphic system prototypes using passive crossbars. Here we report a 64x64 passive metal-oxide memristor crossbar circuit with ~99% device yield, based on a foundry-compatible fabrication process featuring etch-down patterning and low-temperature budget, conducive to vertical integration. The achieved ~26% variations of switching voltages of our devices were sufficient for programming 4K-pixel gray-scale patterns with an average tuning error smaller than 4%. The analog properties were further verified by experimentally demonstrating MNIST pattern classification with a fidelity close to the software-modeled limit for a network of this size, with an ~1% average error of import of ex-situ-calculated synaptic weights. We believe that our work is a significant improvement over the state-of-the-art passive crossbar memories in both complexity and analog properties.