Segmentation: U-Net, Mask R-CNN, and Medical Applications
Connections: Log Likelihood, Cross Entropy, KL Divergence, Logistic Regression, and Neural Networks
Guided Grad-CAM is Broken! Sanity Checks for Saliency Maps
CNN Heat Maps: Gradients vs. DeconvNets vs. Guided Backpropagation
Best Use of Train/Val/Test Splits, with Tips for Medical Data
