Safety Monitoring of Machine Learning Perception Functions: a Survey
Machine Learning (ML) models, such as deep neural networks, are widely applied in
autonomous systems to perform complex perception tasks. New dependability challenges …
autonomous systems to perform complex perception tasks. New dependability challenges …
Unveiling AI's Blind Spots: An Oracle for In-Domain, Out-of-Domain, and Adversarial Errors
S Han, M Zhang - arXiv preprint arXiv:2410.02384, 2024 - arxiv.org
AI models make mistakes when recognizing images-whether in-domain, out-of-domain, or
adversarial. Predicting these errors is critical for improving system reliability, reducing costly …
adversarial. Predicting these errors is critical for improving system reliability, reducing costly …
Evaluating Reliability in Medical DNNs: A Critical Analysis of Feature and Confidence-Based OOD Detection
H Anthony, K Kamnitsas - International Workshop on Uncertainty for Safe …, 2024 - Springer
Reliable use of deep neural networks (DNNs) for medical image analysis requires methods
to identify inputs that differ significantly from the training data, called out-of-distribution …
to identify inputs that differ significantly from the training data, called out-of-distribution …
Can we Defend Against the Unknown? An Empirical Study About Threshold Selection for Neural Network Monitoring
With the increasing use of neural networks in critical systems, runtime monitoring becomes
essential to reject unsafe predictions during inference. Various techniques have emerged to …
essential to reject unsafe predictions during inference. Various techniques have emerged to …
[PDF][PDF] Test and validation of perception-based ADAS: modern solutions to traditional challenges
RS Ferreira - researchgate.net
Perception-based advanced driver-assistance systems (ADAS) are widely applied in
modern vehicles to assist drivers by increasing the vehicle's safe operation. This perception …
modern vehicles to assist drivers by increasing the vehicle's safe operation. This perception …