Abstract: We propose an algorithm combining calibrated prediction and generalization bounds from learning theory to construct confidence sets for deep neural networks with PAC guarantees---i.e., the confidence set for a given input contains the true label with high probability. We demonstrate how our approach can be used to construct PAC confidence sets on ResNet for ImageNet, a visual object tracking model, and a dynamics model for the half-cheetah reinforcement learning problem.

Similar Papers

Distance-Based Learning from Errors for Confidence Calibration
Chen Xing, Sercan Arik, Zizhao Zhang, Tomas Pfister,