Gradient descent-based programming of analog in-memory computing cores

Abstract

The precise programming of crossbar arrays of unit-cells is crucial for obtaining high matrix-vector-multiplication (MVM) accuracy in analog in-memory computing (AIMC) cores. We propose a radically different approach based on directly minimizing the MVM error using gradient descent with synthetic random input data. Our method significantly reduces the MVM error compared with conventional unit-cell by unit-cell iterative programming. It also eliminates the need for high-resolution analog-to-digital converters (ADCs) to read the small unit-cell conductance during programming. Our method improves the experimental inference accuracy of ResNet-9 implemented on two phase-change memory (PCM)-based AIMC cores by 1.26%.

Cited by in corpus (5)

• Fast offset corrected in-memory training
• HyDe: A Hybrid PCM/FeFET/SRAM Device-search for Optimizing Area and Energy-efficiencies in Analog IMC Platforms
• Improving the Accuracy of Analog-Based In-Memory Computing Accelerators Post-Training
• In-Materia Speech Recognition
• A Precision-Optimized Fixed-Point Near-Memory Digital Processing Unit for Analog In-Memory Computing