# Discovering and Explaining the Non-Causality of Deep Learning in SAR ATR

arXiv: https://arxiv.org/abs/2304.00668 · HTML: https://mlsanity.com/paper/2304.00668/

## Key facts

- Cited by 9 papers in the map (data through 14 Oct 2026).
- Builds on 12 of its 36 references that are in the map.
- First posted on arXiv 3 Apr 2023.

## Builds on (most cited first, 12 of 12)

- [Deep Residual Learning for Image Recognition](https://mlsanity.com/paper/1512.03385/) (2015)
- [A Unified Approach to Interpreting Model Predictions](https://mlsanity.com/paper/1705.07874/) (2017)
- [EfficientNet: Rethinking Model Scaling for Convolutional Neural Networks](https://mlsanity.com/paper/1905.11946/) (2019)
- [A ConvNet for the 2020s](https://mlsanity.com/paper/2201.03545/) (2022)
- [The many Shapley values for model explanation](https://mlsanity.com/paper/1908.08474/) (2019)
- [Approximating CNNs with Bag-of-local-Features models works surprisingly well on ImageNet](https://mlsanity.com/paper/1904.00760/) (2019)
- [A Survey on Causal Inference](https://mlsanity.com/paper/2002.02770/) (2020)
- [Informative Dropout for Robust Representation Learning: A Shape-bias Perspective](https://mlsanity.com/paper/2008.04254/) (2020)
- [From BoW to CNN: Two Decades of Texture Representation for Texture Classification](https://mlsanity.com/paper/1801.10324/) (2018)
- [Examining CNN Representations with respect to Dataset Bias](https://mlsanity.com/paper/1710.10577/) (2017)
- [Automatic Target Recognition on Synthetic Aperture Radar Imagery: A Survey](https://mlsanity.com/paper/2007.02106/) (2020)
- [Scattering Model Guided Adversarial Examples for SAR Target Recognition: Attack and Defense](https://mlsanity.com/paper/2209.04779/) (2022)

## Built on by (most cited first, 9 of 9)

- [SARATR-X: Toward Building A Foundation Model for SAR Target Recognition](https://mlsanity.com/paper/2405.09365/) (2024)
- [ATRNet-STAR: A Large Dataset and Benchmark Towards Remote Sensing Object Recognition in the Wild](https://mlsanity.com/paper/2501.13354/) (2025)
- [Hierarchical Disentanglement-Alignment Network for Robust SAR Vehicle Recognition](https://mlsanity.com/paper/2304.03550/) (2023)
- [IRASNet: Improved Feature-Level Clutter Reduction for Domain Generalized SAR-ATR](https://mlsanity.com/paper/2409.16845/) (2024)
- [Learning Invariant Representation via Contrastive Feature Alignment for Clutter Robust SAR Target Recognition](https://mlsanity.com/paper/2304.01747/) (2023)
- [Multi-task SAR Image Processing via GAN-based Unsupervised Manipulation](https://mlsanity.com/paper/2408.01553/) (2024)
- [Studying the Effects of Self-Attention on SAR Automatic Target Recognition](https://mlsanity.com/paper/2409.00473/) (2024)
- [Progressive Multi-Level Alignments for Semi-Supervised Domain Adaptation SAR Target Recognition Using Simulated Data](https://mlsanity.com/paper/2411.04711/) (2024)
- [SRAW-Attack: Space-Reweighted Adversarial Warping Attack for SAR Target Recognition](https://mlsanity.com/paper/2601.10324/) (2026)

## Abstract (arXiv)

In recent years, deep learning has been widely used in SAR ATR and achieved excellent performance on the MSTAR dataset. However, due to constrained imaging conditions, MSTAR has data biases such as background correlation, i.e., background clutter properties have a spurious correlation with target classes. Deep learning can overfit clutter to reduce training errors. Therefore, the degree of overfitting for clutter reflects the non-causality of deep learning in SAR ATR. Existing methods only qualitatively analyze this phenomenon. In this paper, we quantify the contributions of different regions to target recognition based on the Shapley value. The Shapley value of clutter measures the degree of overfitting. Moreover, we explain how data bias and model bias contribute to non-causality. Concisely, data bias leads to comparable signal-to-clutter ratios and clutter textures in training and test sets. And various model structures have different degrees of overfitting for these biases. The experimental results of various models under standard operating conditions on the MSTAR dataset support our conclusions. Our code is available at https://github.com/waterdisappear/Data-Bias-in-MSTAR.