# Improving Uncertainty of Deep Learning-based Object Classification on Radar Spectra using Label Smoothing

arXiv: https://arxiv.org/abs/2109.12851 · HTML: https://mlsanity.com/paper/2109.12851/

## Key facts

- Cited by 4 papers in the map (data through 14 Oct 2026).
- Builds on 15 of its 31 references that are in the map.
- First posted on arXiv 27 Sep 2021.

## Builds on (most cited first, 15 of 15)

- [Distilling the Knowledge in a Neural Network](https://mlsanity.com/paper/1503.02531/) (2015)
- [Rethinking the Inception Architecture for Computer Vision](https://mlsanity.com/paper/1512.00567/) (2015)
- [mixup: Beyond Empirical Risk Minimization](https://mlsanity.com/paper/1710.09412/) (2017)
- [On Calibration of Modern Neural Networks](https://mlsanity.com/paper/1706.04599/) (2017)
- [When Does Label Smoothing Help?](https://mlsanity.com/paper/1906.02629/) (2019)
- [Why ReLU networks yield high-confidence predictions far away from the training data and how to mitigate the problem](https://mlsanity.com/paper/1812.05720/) (2018)
- [On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks](https://mlsanity.com/paper/1905.11001/) (2019)
- [Sampling-free Epistemic Uncertainty Estimation Using Approximated Variance Propagation](https://mlsanity.com/paper/1908.00598/) (2019)
- [RODNet: A Real-Time Radar Object Detection Network Cross-Supervised by Camera-Radar Fused Object 3D Localization](https://mlsanity.com/paper/2102.05150/) (2021)
- [CNN based Road User Detection using the 3D Radar Cube](https://mlsanity.com/paper/2004.12165/) (2020)
- [Multi-Class Uncertainty Calibration via Mutual Information Maximization-based Binning](https://mlsanity.com/paper/2006.13092/) (2020)
- [On-manifold Adversarial Data Augmentation Improves Uncertainty Calibration](https://mlsanity.com/paper/1912.07458/) (2019)
- [Using Machine Learning to Detect Ghost Images in Automotive Radar](https://mlsanity.com/paper/2007.05280/) (2020)
- [DeepReflecs: Deep Learning for Automotive Object Classification with Radar Reflections](https://mlsanity.com/paper/2010.09273/) (2020)
- [Investigation of Uncertainty of Deep Learning-based Object Classification on Radar Spectra](https://mlsanity.com/paper/2106.05870/) (2021)

## Built on by (most cited first, 4 of 4)

- [Radars for Autonomous Driving: A Review of Deep Learning Methods and Challenges](https://mlsanity.com/paper/2306.09304/) (2023)
- [A recurrent CNN for online object detection on raw radar frames](https://mlsanity.com/paper/2212.11172/) (2022)
- [Calibrating the Rigged Lottery: Making All Tickets Reliable](https://mlsanity.com/paper/2302.09369/) (2023)
- [Embracing Unknown Step by Step: Towards Reliable Sparse Training in Real World](https://mlsanity.com/paper/2403.20047/) (2024)

## Abstract (arXiv)

Object type classification for automotive radar has greatly improved with recent deep learning (DL) solutions, however these developments have mostly focused on the classification accuracy. Before employing DL solutions in safety-critical applications, such as automated driving, an indispensable prerequisite is the accurate quantification of the classifiers' reliability. Unfortunately, DL classifiers are characterized as black-box systems which output severely over-confident predictions, leading downstream decision-making systems to false conclusions with possibly catastrophic consequences. We find that deep radar classifiers maintain high-confidences for ambiguous, difficult samples, e.g. small objects measured at large distances, under domain shift and signal corruptions, regardless of the correctness of the predictions. The focus of this article is to learn deep radar spectra classifiers which offer robust real-time uncertainty estimates using label smoothing during training. Label smoothing is a technique of refining, or softening, the hard labels typically available in classification datasets. In this article, we exploit radar-specific know-how to define soft labels which encourage the classifiers to learn to output high-quality calibrated uncertainty estimates, thereby partially resolving the problem of over-confidence. Our investigations show how simple radar knowledge can easily be combined with complex data-driven learning algorithms to yield safe automotive radar perception.