Robust Deep Learning for Autonomous Driving
This work addresses safety concerns in autonomous driving by improving uncertainty estimation, but it is incremental as it builds on existing methods for confidence measures and domain adaptation.
The thesis tackled the problem of unreliable uncertainty estimates in deep neural networks for autonomous driving by introducing a new confidence criterion called true class probability (TCP) and a learning scheme to estimate it, validated on image classification and segmentation datasets, and extended to domain adaptation and misclassification detection.
The last decade's research in artificial intelligence had a significant impact on the advance of autonomous driving. Yet, safety remains a major concern when it comes to deploying such systems in high-risk environments. The objective of this thesis is to develop methodological tools which provide reliable uncertainty estimates for deep neural networks. First, we introduce a new criterion to reliably estimate model confidence: the true class probability (TCP). We show that TCP offers better properties for failure prediction than current uncertainty measures. Since the true class is by essence unknown at test time, we propose to learn TCP criterion from data with an auxiliary model, introducing a specific learning scheme adapted to this context. The relevance of the proposed approach is validated on image classification and semantic segmentation datasets. Then, we extend our learned confidence approach to the task of domain adaptation where it improves the selection of pseudo-labels in self-training methods. Finally, we tackle the challenge of jointly detecting misclassification and out-of-distributions samples by introducing a new uncertainty measure based on evidential models and defined on the simplex.