Practical Assessment of Generalization Performance Robustness for Deep Networks via Contrastive Examples
This work addresses the need for more reliable generalization assessment in deep learning, offering a practical tool for researchers and practitioners, though it is incremental as it builds on existing contrastive learning ideas.
The authors tackled the problem of evaluating generalization performance in deep neural networks by proposing ContRE, a framework that uses contrastive examples from data transformations to assess models, finding strong correlations with testing set performance and confirming its robustness as a complementary measure.
Training images with data transformations have been suggested as contrastive examples to complement the testing set for generalization performance evaluation of deep neural networks (DNNs). In this work, we propose a practical framework ContRE (The word "contre" means "against" or "versus" in French.) that uses Contrastive examples for DNN geneRalization performance Estimation. Specifically, ContRE follows the assumption in contrastive learning that robust DNN models with good generalization performance are capable of extracting a consistent set of features and making consistent predictions from the same image under varying data transformations. Incorporating with a set of randomized strategies for well-designed data transformations over the training set, ContRE adopts classification errors and Fisher ratios on the generated contrastive examples to assess and analyze the generalization performance of deep models in complement with a testing set. To show the effectiveness and the efficiency of ContRE, extensive experiments have been done using various DNN models on three open source benchmark datasets with thorough ablation studies and applicability analyses. Our experiment results confirm that (1) behaviors of deep models on contrastive examples are strongly correlated to what on the testing set, and (2) ContRE is a robust measure of generalization performance complementing to the testing set in various settings.