ROSAR: An Adversarial Re-Training Framework for Robust Side-Scan Sonar Object DetectionMartin Aubard, László Antal, Ana Madureira et al.
This paper introduces ROSAR, a novel framework enhancing the robustness of deep learning object detection models tailored for side-scan sonar (SSS) images, generated by autonomous underwater vehicles using sonar sensors. By extending our prior work on knowledge distillation (KD), this framework integrates KD with adversarial retraining to address the dual challenges of model efficiency and robustness against SSS noises. We introduce three novel, publicly available SSS datasets, capturing different sonar setups and noise conditions. We propose and formalize two SSS safety properties and utilize them to generate adversarial datasets for retraining. Through a comparative analysis of projected gradient descent (PGD) and patch-based adversarial attacks, ROSAR demonstrates significant improvements in model robustness and detection accuracy under SSS-specific conditions, enhancing the model's robustness by up to 1.85%. ROSAR is available at https://github.com/remaro-network/ROSAR-framework.
Knowledge Distillation in YOLOX-ViT for Side-Scan Sonar Object DetectionMartin Aubard, László Antal, Ana Madureira et al.
In this paper we present YOLOX-ViT, a novel object detection model, and investigate the efficacy of knowledge distillation for model size reduction without sacrificing performance. Focused on underwater robotics, our research addresses key questions about the viability of smaller models and the impact of the visual transformer layer in YOLOX. Furthermore, we introduce a new side-scan sonar image dataset, and use it to evaluate our object detector's performance. Results show that knowledge distillation effectively reduces false positives in wall detection. Additionally, the introduced visual transformer layer significantly improves object detection accuracy in the underwater environment. The source code of the knowledge distillation in the YOLOX-ViT is at https://github.com/remaro-network/KD-YOLOX-ViT.
2.0LGNov 16, 2023
Extending Neural Network Verification to a Larger Family of Piece-wise Linear Activation FunctionsLászló Antal, Hana Masara, Erika Ábrahám
In this paper, we extend an available neural network verification technique to support a wider class of piece-wise linear activation functions. Furthermore, we extend the algorithms, which provide in their original form exact respectively over-approximative results for bounded input sets represented as start sets, to allow also unbounded input set. We implemented our algorithms and demonstrated their effectiveness in some case studies.
6.4LGJun 18
Shifting-based Optimizable Linear Relaxations for General Activation FunctionsPhilipp Kern, László Antal, Erika Ábráham et al.
The use of neural networks (NNs) is rapidly increasing, including in safety- and security-critical domains. To provide formal guarantees about NN behavior, many verification methods rely on optimizable linear relaxations of activation functions. However, existing techniques depend on hand-crafted relaxations for each activation function. Extension to state-of-the-art activation functions therefore requires substantial manual effort. In contrast, our approach SLiR (Shifting-based Linear Relaxations) is broadly applicable, requiring only a Lipschitz constant or a set of critical points. SLiR parameterizes relaxations by their slope and computes the corresponding offset via a shifting procedure that ensures sound upper and lower bounds over the input domain, enabling efficient optimization while maintaining correctness. Our experiments show that SLiR produces tight relaxations across a wide range of practical activation functions and enables verification of up to 7.8x more properties compared to state-of-the-art methods.