Title: Consistent Relative Confidence and Label-Free Model Selection for Convolutional Neural Networks

Abstract: In this paper, we are concerned with image classification with deep convolutional neural networks (CNNs). We focus on the following question: given a set of candidate CNN models, how to select the right one with the best generalization property for the current task? Current model selection methods all require access to a batch of labeled data for computing a pre-specified performance metric, such as the cross-entropy loss, the classification error rate and the negative log-likelihood. In many practical cases, labels are not available in time as labeling itself is a time-consuming and expensive task. To this end, we propose an approach to CNN model selection using only unlabeled data. We develop this method based on a principle termed consistent relative confidence. Experimental results on benchmark datasets demonstrate the effectiveness and efficiency of our method.