mlsanityData to 7 Oct 2026

Sensitivity-Aware Finetuning for Accuracy Recovery on Deep Learning Hardware

arXiv 2306.03076 ↗First posted 5 Jun 2023cs.LG cs.ARcited by 1

Not on the map. Why?

Abstract (arXiv)

Existing methods to recover model accuracy on analog-digital hardware in the presence of quantization and analog noise include noise-injection training. However, it can be slow in practice, incurring high computational costs, even when starting from pretrained models. We introduce the Sensitivity-Aware Finetuning (SAFT) approach that identifies noise sensitive layers in a model, and uses the information to freeze specific layers for noise-injection training. Our results show that SAFT achieves comparable accuracy to noise-injection training and is 2x to 8x faster.

Key facts. Cited by 1 paper in the map (data through 7 Oct 2026). Builds on 13 of its 18 references that are in the map. First posted on arXiv 5 Jun 2023.

Builds on

13 of 18 references.

References · 13

  1. 2021BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-modelscited by 1,234
  2. 2019A Comprehensive Survey on Transfer Learningcited by 1,154
  3. 2016XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networkscited by 829
  4. 2021A Survey of Quantization Methods for Efficient Neural Network Inferencecited by 798
  5. 2018Tune: A Research Platform for Distributed Model Selection and Trainingcited by 395
  6. 2020Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluationcited by 183
  7. 2018Discovering Low-Precision Networks Close to Full-Precision Networks for Efficient Embedded Inferencecited by 33
  8. 2022Efficient Fine-Tuning of BERT Models on the Edgecited by 29
  9. 2020Noisy Machines: Understanding Noisy Neural Networks and Enhancing Robustness to Analog Hardware Errors Using Distillationcited by 17
  10. 2021Dynamic Precision Analog Computing for Neural Networkscited by 12
  11. 2018UNIQ: Uniform Noise Injection for Non-Uniform Quantization of Neural Networkscited by 9
  12. 2021Adaptive Precision Training (AdaPT): A dynamic fixed point quantized training approach for DNNscited by 8
1 more
  1. 2022Adaptive Block Floating-Point for Analog Deep Learning Hardwarecited by 4

Built on by

Cited by year (2026 to 7 Oct)

Citing papers per year: 2026: 1.

1 paper.

  1. 2026Quantum Tunneling-Aware Machine Learning: Physics-Derived Noise Models for Robust Deploymentcited by 0