English

Improving the Accuracy of Analog-Based In-Memory Computing Accelerators Post-Training

Emerging Technologies 2025-01-30 v1

Abstract

Analog-Based In-Memory Computing (AIMC) inference accelerators can be used to efficiently execute Deep Neural Network (DNN) inference workloads. However, to mitigate accuracy losses, due to circuit and device non-idealities, Hardware-Aware (HWA) training methodologies must be employed. These typically require significant information about the underlying hardware. In this paper, we propose two Post-Training (PT) optimization methods to improve accuracy after training is performed. For each crossbar, the first optimizes the conductance range of each column, and the second optimizes the input, i.e, Digital-to-Analog Converter (DAC), range. It is demonstrated that, when these methods are employed, the complexity during training, and the amount of information about the underlying hardware can be reduced, with no notable change in accuracy (\leq0.1%) when finetuning the pretrained RoBERTa transformer model for all General Language Understanding Evaluation (GLUE) benchmark tasks. Additionally, it is demonstrated that further optimizing learned parameters PT improves accuracy.

Keywords

Cite

@article{arxiv.2401.09859,
  title  = {Improving the Accuracy of Analog-Based In-Memory Computing Accelerators Post-Training},
  author = {Corey Lammie and Athanasios Vasilopoulos and Julian Büchel and Giacomo Camposampiero and Manuel Le Gallo and Malte Rasch and Abu Sebastian},
  journal= {arXiv preprint arXiv:2401.09859},
  year   = {2025}
}

Comments

Accepted at 2024 IEEE International Symposium on Circuits and Systems (ISCAS)

R2 v1 2026-06-28T14:20:13.501Z