EPIM: Efficient Processing-In-Memory Accelerators based on Epitome

arXiv:2311.07620v2 Announce Type: replace-cross
Abstract: The utilization of large-scale neural networks on Processing-In-Memory (PIM) accelerators encounters challenges due to constrained on-chip memory capacity. To tackle this issue, current works explore model compression algorithms to reduce the size of Convolutional Neural Networks (CNNs). Most of these algorithms either aim to represent neural operators with reduced-size parameters (e.g., quantization) or search for the best combinations of neural operators (e.g., neural architecture search). Designing neural operators to align with PIM accelerators' specifications is …

abstract accelerators aim algorithms arxiv capacity challenges chip cnns compression convolutional neural networks cs.ar cs.lg current explore in-memory issue memory networks neural networks operators processing reduce scale type