A Device Non-Ideality Resilient Approach for Mapping Neural Networks to Crossbar Arrays
Arman Kazemi, Cristobal Alessandri, Alan C. Seabaugh, Xiaobo Sharon Hu, Michael T. Niemier, Siddharth Joshi
Abstract
We propose a technology-independent method, referred to as adjacent connection matrix (ACM), to efficiently map signed weight matrices to non-negative crossbar arrays. When compared to same-hardware-overhead mapping methods, using ACM leads to improvements of up to 20% in training accuracy for ResNet-20 with the CIFAR-10 dataset when training with 5-bit precision crossbar arrays or lower. When compared with strategies that use two elements to represent a weight, ACM achieves comparable training accuracies, while also offering area and read energy reductions of 2.3× and 7×, respectively. ACM also has a mild regularization effect that improves inference accuracy in crossbar arrays without any retraining or costly device/variation-aware training.
Ask Lune
Know this paper in minutes, not an afternoon.
Lune can read every page of it and the top-tier papers that cite it, and gives you its claims, its evidence and its limits, each one cited to the paper it comes from. No skimming, no AI slop.
Related papers
- InfoX: an energy-efficient ReRAM accelerator design with information-lossless low-bit ADCsYintao He, Songyun Qu, Ying Wang, Bing Li et al.DAC 2022 · 10 citations
- PIM-Prune: Fine-Grain DCNN Pruning for Crossbar-Based Process-In-Memory ArchitectureChaoqun Chu, Yanzhi Wang, Yilong Zhao, Xiaolong Ma et al.DAC 2020 · 64 citations
- Effective zero compression on ReRAM-based sparse DNN acceleratorsHoon Shin, Rihae Park, Seung Yul Lee, Yeonhong Park et al.DAC 2022 · 11 citations
- High Energy-efficiency and Low latency In-Memory Computing using Analog Accumulator and In-Memory ADC with shared ReferencesJunyi Yang, Shuai Dong, Zhengnan Fu, Hongyang Shang et al.DAC 2025 · 3 citations
- Towards resilient analog in-memory deep learning via data layout re-organizationMuhammad Rashedul Haq Rashed, Amro Awad, Sumit Kumar Jha, Rickard EwetzDAC 2022 · 2 citations
