1

Repairing Neural Networks by Leaving the Right Past Behind

Uncertainty Quantification in Medical Image Synthesis

Learning to Downsample for Segmentation of Ultra-High Resolution Images

A Principled Approach to Failure Analysis and Model Repairment: Demonstration in Medical Imaging

Active Label Cleaning for Improved Dataset Quality under Resource Constraints

Disentangling Human Error from the Ground Truth in Segmentation of Medical Images

Foveation for Segmentation of Ultra-High Resolution Images

Uncertainty Quantification in Deep Learning for Safer Neuroimage Enhancement

Stochastic Filter Groups for Multi-Task CNNs: Learning Specialist and Generalist Convolution Kernels

Adaptive Neural Trees