# Sensitivity-Aware Finetuning for Accuracy Recovery on Deep Learning Hardware

arXiv: https://arxiv.org/abs/2306.03076 · HTML: https://mlsanity.com/paper/2306.03076/

## Key facts

- Cited by 1 papers in the map (data through 14 Oct 2026).
- Builds on 13 of its 18 references that are in the map.
- First posted on arXiv 5 Jun 2023.

## Builds on (most cited first, 13 of 13)

- [BitFit: Simple Parameter-efficient Fine-tuning for Transformer-based Masked Language-models](https://mlsanity.com/paper/2106.10199/) (2021)
- [A Comprehensive Survey on Transfer Learning](https://mlsanity.com/paper/1911.02685/) (2019)
- [XNOR-Net: ImageNet Classification Using Binary Convolutional Neural Networks](https://mlsanity.com/paper/1603.05279/) (2016)
- [A Survey of Quantization Methods for Efficient Neural Network Inference](https://mlsanity.com/paper/2103.13630/) (2021)
- [Tune: A Research Platform for Distributed Model Selection and Training](https://mlsanity.com/paper/1807.05118/) (2018)
- [Integer Quantization for Deep Learning Inference: Principles and Empirical Evaluation](https://mlsanity.com/paper/2004.09602/) (2020)
- [Discovering Low-Precision Networks Close to Full-Precision Networks for Efficient Embedded Inference](https://mlsanity.com/paper/1809.04191/) (2018)
- [Efficient Fine-Tuning of BERT Models on the Edge](https://mlsanity.com/paper/2205.01541/) (2022)
- [Noisy Machines: Understanding Noisy Neural Networks and Enhancing Robustness to Analog Hardware Errors Using Distillation](https://mlsanity.com/paper/2001.04974/) (2020)
- [Dynamic Precision Analog Computing for Neural Networks](https://mlsanity.com/paper/2102.06365/) (2021)
- [UNIQ: Uniform Noise Injection for Non-Uniform Quantization of Neural Networks](https://mlsanity.com/paper/1804.10969/) (2018)
- [Adaptive Precision Training (AdaPT): A dynamic fixed point quantized training approach for DNNs](https://mlsanity.com/paper/2107.13490/) (2021)
- [Adaptive Block Floating-Point for Analog Deep Learning Hardware](https://mlsanity.com/paper/2205.06287/) (2022)

## Built on by (most cited first, 1 of 1)

- [Quantum Tunneling-Aware Machine Learning: Physics-Derived Noise Models for Robust Deployment](https://mlsanity.com/paper/2606.00741/) (2026)

## 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.