Research paper proposing a novel method for uncertainty quantification in convolutional neural networks using bootstrap approaches with convex neural network models. Applies ensemble-based uncertainty estimation techniques to improve model reliability and interpretability.
Research
Uncertainty Quantification in CNN Through the Bootstrap of Convex Neural Networks
Bootstrap convex neural networks enable quantifiable uncertainty in CNNs, making models more interpretable and reliable for high-stakes applications.
Wednesday, April 15, 2026 12:00 PM UTC2 MIN READSOURCE: arXiv CS.LG (Machine Learning)BY sys://pipeline
Tags
research