# Stabilizing Information Flow Entropy: Regularization for Safe and Interpretable Autonomous Driving Perception

arXiv: https://arxiv.org/abs/2509.16277 · HTML: https://mlsanity.com/paper/2509.16277/

## Key facts

- No citing papers in the map yet (data through 14 Oct 2026).
- Builds on 10 of its 28 references that are in the map.
- First posted on arXiv 18 Sep 2025.

## Builds on (most cited first, 10 of 10)

- [What Uncertainties Do We Need in Bayesian Deep Learning for Computer Vision?](https://mlsanity.com/paper/1703.04977/) (2017)
- [YOLOv9: Learning What You Want to Learn Using Programmable Gradient Information](https://mlsanity.com/paper/2402.13616/) (2024)
- [DeepJSCC-f: Deep Joint Source-Channel Coding of Images with Feedback](https://mlsanity.com/paper/1911.11174/) (2019)
- [Analysis of Explainers of Black Box Deep Neural Networks for Computer Vision: A Survey](https://mlsanity.com/paper/1911.12116/) (2019)
- [On the Practicality of Deterministic Epistemic Uncertainty](https://mlsanity.com/paper/2107.00649/) (2021)
- [Multi-modal Graph Learning for Disease Prediction](https://mlsanity.com/paper/2203.05880/) (2022)
- [Achieving Robustness to Aleatoric Uncertainty with Heteroscedastic Bayesian Optimisation](https://mlsanity.com/paper/1910.07779/) (2019)
- [Deep Joint Source-Channel Coding for Wireless Image Retrieval](https://mlsanity.com/paper/1910.12703/) (2019)
- [On the Robustness of Monte Carlo Dropout Trained with Noisy Labels](https://mlsanity.com/paper/2103.12002/) (2021)
- [A novel multimodal fusion network based on a joint coding model for lane line segmentation](https://mlsanity.com/paper/2103.11114/) (2021)

## Abstract (arXiv)

Deep perception networks in autonomous driving traditionally rely on data-intensive training regimes and post-hoc anomaly detection, often disregarding fundamental information-theoretic constraints governing stable information processing. We reconceptualize deep neural encoders as hierarchical communication chains that incrementally compress raw sensory inputs into task-relevant latent features. Within this framework, we establish two theoretically justified design principles for robust perception: (D1) smooth variation of mutual information between consecutive layers, and (D2) monotonic decay of latent entropy with network depth. Our analysis shows that, under realistic architectural assumptions, particularly blocks comprising repeated layers of similar capacity, enforcing smooth information flow (D1) naturally encourages entropy decay (D2), thus ensuring stable compression. Guided by these insights, we propose Eloss, a novel entropy-based regularizer designed as a lightweight, plug-and-play training objective. Rather than marginal accuracy improvements, this approach represents a conceptual shift: it unifies information-theoretic stability with standard perception tasks, enabling explicit, principled detection of anomalous sensor inputs through entropy deviations. Experimental validation on large-scale 3D object detection benchmarks (KITTI and nuScenes) demonstrates that incorporating Eloss consistently achieves competitive or improved accuracy while dramatically enhancing sensitivity to anomalies, amplifying distribution-shift signals by up to two orders of magnitude. This stable information-compression perspective not only improves interpretability but also establishes a solid theoretical foundation for safer, more robust autonomous driving perception systems.