Fetching the paper…

On Mixup Training: Improved Calibration and Predictive Uncertainty for Deep Neural Networks · Around