# Maximum Likelihood Distillation for Robust Modulation Classification

arXiv: https://arxiv.org/abs/2211.00748 · HTML: https://mlsanity.com/paper/2211.00748/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 24 of its 38 references that are in the map.
- First posted on arXiv 1 Nov 2022.

## Builds on (most cited first, 24 of 24)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Explaining and Harnessing Adversarial Examples](https://mlsanity.com/paper/1412.6572/) (2014)
- [Towards Deep Learning Models Resistant to Adversarial Attacks](https://mlsanity.com/paper/1706.06083/) (2017)
- [Intriguing properties of neural networks](https://mlsanity.com/paper/1312.6199/) (2013)
- [FitNets: Hints for Thin Deep Nets](https://mlsanity.com/paper/1412.6550/) (2014)
- [Theoretically Principled Trade-off between Robustness and Accuracy](https://mlsanity.com/paper/1901.08573/) (2019)
- [Inception-v4, Inception-ResNet and the Impact of Residual Connections on Learning](https://mlsanity.com/paper/1602.07261/) (2016)
- [Paying More Attention to Attention: Improving the Performance of Convolutional Neural Networks via Attention Transfer](https://mlsanity.com/paper/1612.03928/) (2016)
- [Universal adversarial perturbations](https://mlsanity.com/paper/1610.08401/) (2016)
- [Adversarially Robust Distillation](https://mlsanity.com/paper/1905.09747/) (2019)
- [Over the Air Deep Learning Based Radio Signal Classification](https://mlsanity.com/paper/1712.04578/) (2017)
- [Convolutional Radio Modulation Recognition Networks](https://mlsanity.com/paper/1602.04105/) (2016)
- [Revisiting Adversarial Robustness Distillation: Robust Soft Labels Make Student Better](https://mlsanity.com/paper/2108.07969/) (2021)
- [Adversarial Attacks on Deep-Learning Based Radio Signal Classification](https://mlsanity.com/paper/1808.07713/) (2018)
- [Distributed Deep Learning Models for Wireless Signal Classification with Low-Cost Spectrum Sensors](https://mlsanity.com/paper/1707.08908/) (2017)
- [How benign is benign overfitting?](https://mlsanity.com/paper/2007.04028/) (2020)
- [Deep Architectures for Modulation Recognition](https://mlsanity.com/paper/1703.09197/) (2017)
- [Evaluating Adversarial Evasion Attacks in the Context of Wireless Communications](https://mlsanity.com/paper/1903.01563/) (2019)
- [Channel-Aware Adversarial Attacks Against Deep Learning-Based Wireless Signal Classifiers](https://mlsanity.com/paper/2005.05321/) (2020)
- [On the benefits of knowledge distillation for adversarial robustness](https://mlsanity.com/paper/2203.07159/) (2022)
- [How and When Adversarial Robustness Transfers in Knowledge Distillation?](https://mlsanity.com/paper/2110.12072/) (2021)
- [Open Set Modulation Recognition Based on Dual-Channel LSTM Model](https://mlsanity.com/paper/2002.12037/) (2020)
- [On the benefits of robust models in modulation recognition](https://mlsanity.com/paper/2103.14977/) (2021)
- [SafeAMC: Adversarial training for robust modulation recognition models](https://mlsanity.com/paper/2105.13746/) (2021)

## Abstract (arXiv)

Deep Neural Networks are being extensively used in communication systems and Automatic Modulation Classification (AMC) in particular. However, they are very susceptible to small adversarial perturbations that are carefully crafted to change the network decision. In this work, we build on knowledge distillation ideas and adversarial training in order to build more robust AMC systems. We first outline the importance of the quality of the training data in terms of accuracy and robustness of the model. We then propose to use the Maximum Likelihood function, which could solve the AMC problem in offline settings, to generate better training labels. Those labels teach the model to be uncertain in challenging conditions, which permits to increase the accuracy, as well as the robustness of the model when combined with adversarial training. Interestingly, we observe that this increase in performance transfers to online settings, where the Maximum Likelihood function cannot be used in practice. Overall, this work highlights the potential of learning to be uncertain in difficult scenarios, compared to directly removing label noise.