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

Abstract

This paper is concerned with image classification based on deep convolutional neural networks (CNNs). The focus is centered around the following question: given a set of candidate CNN models, how to select the right one that has the best generalization property for the current task? Present model selection methods require access to a batch of labeled data for defining a performance metric, such as the cross-entropy loss, the classification error rate, the negative log-likelihood, and so on. In many practical cases, however, labeled data are not available in time as labeling itself is a time-consuming and expensive task. To this end, this paper presents an approach to CNN model selection using only unlabeled data. This method is developed based on a principle termed consistent relative confidence (CRC). The effectiveness and efficiency of the presented method are demonstrated by extensive experimental studies based on datasets MNIST and FasionM-NIST.

Enabling Deep Learning for All-in EDGE paradigm

“…Different metrics are selected based on the type of problem, i.e., classification or regression. Some of the widely used loss functions to capture the learning of DNN at EDGE while training are Mean Absolute Error [60,64,273], Mean Square Error Loss [269,289,316], Negative Log-Likelihood Loss [138,147,219], Cross-Entropy Loss [54,57,128,162], Kullback-Leibler divergence [47,70,237,294] etc.…”