Uncertainty-based Meta-Reinforcement Learning for Robust Radar Tracking
This work addresses robustness issues in radar tracking for real-life applications, though it is incremental as it builds on existing uncertainty and meta-learning techniques.
The paper tackles the problem of limited robustness and distributional shift in deep learning for radar tracking by proposing an uncertainty-based meta-reinforcement learning approach with out-of-distribution detection, resulting in a 16% improvement over related Meta-RL methods and a 35% improvement over the baseline on unseen scenarios, with an F1-Score of 72% for OOD detection.
Nowadays, Deep Learning (DL) methods often overcome the limitations of traditional signal processing approaches. Nevertheless, DL methods are barely applied in real-life applications. This is mainly due to limited robustness and distributional shift between training and test data. To this end, recent work has proposed uncertainty mechanisms to increase their reliability. Besides, meta-learning aims at improving the generalization capability of DL models. By taking advantage of that, this paper proposes an uncertainty-based Meta-Reinforcement Learning (Meta-RL) approach with Out-of-Distribution (OOD) detection. The presented method performs a given task in unseen environments and provides information about its complexity. This is done by determining first and second-order statistics on the estimated reward. Using information about its complexity, the proposed algorithm is able to point out when tracking is reliable. To evaluate the proposed method, we benchmark it on a radar-tracking dataset. There, we show that our method outperforms related Meta-RL approaches on unseen tracking scenarios in peak performance by 16% and the baseline by 35% while detecting OOD data with an F1-Score of 72%. This shows that our method is robust to environmental changes and reliably detects OOD scenarios.