Crossbar-aligned & integer-only neural network compression for efficient in-memory acceleration

• Main Authors: Huai, Shuo, Liu, Di, Luo, Xiangzhong, Chen, Hui, Liu, Weichen, Subramaniam, Ravi
• Other Authors: School of Computer Science and Engineering
• Format: Conference or Workshop Item
• Language: English
• Published: 2023
• Subjects: Engineering::Computer science and engineering::Computing methodologies; In-Memory Computing; Pruning; Quantization; Neural Networks
• Online Access: https://hdl.handle.net/10356/165352
• Institution: Nanyang Technological University

Crossbar-based In-Memory Computing (IMC) accelerators preload the entire Deep Neural Network (DNN) into crossbars before inference. However, devices with limited crossbars cannot infer increasingly complex models. IMC-pruning can reduce the usage of crossbars, but current methods need expensive extr...