Cross-entropy is a widely used loss function in applications. It coincides with the logistic loss
applied to the outputs of a neural network, when the softmax is used. But, what guarantees …

In recent years, there has been an explosion of research into develo** more robust deep
neural networks against adversarial examples. Adversarial training appears as one of the …

Abstract Interpretability of Deep Learning (DL) is a barrier to trustworthy AI. Despite great
efforts made by the Explainable AI (XAI) community, explanations lack robustness …

Deep neural networks are susceptible to adversarial attacks due to the accumulation of
perturbations in the feature level, and numerous works have boosted model robustness by …

Increasing numbers of artificial intelligence systems are employing collaborative machine
learning techniques, such as federated learning, to build a shared powerful deep model …

Cooperative multi-agent reinforcement learning (c-MARL) is widely applied in safety-critical
scenarios, thus the analysis of robustness for c-MARL models is profoundly important …

Abstract 3D point cloud models are widely applied in safety-critical scenes, which delivers
an urgent need to obtain more solid proofs to verify the robustness of models. Existing …

Robust pedestrian trajectory forecasting is crucial to develo** safe autonomous vehicles.
Although previous works have studied adversarial robustness in the context of trajectory …

Formal methods and machine learning are two research fields with drastically different
foundations and philosophies. Formal methods utilise mathematically rigorous techniques …

Deep neural networks are vulnerable to adversarial examples, which are well-designed
examples aiming to cause models to produce wrong outputs with high confidence. Although …