Non-adversarial Robustness of Deep Learning Methods for Computer Vision
It addresses the challenge of maintaining model performance under natural distribution shifts in computer vision, but is incremental as it reviews existing approaches.
This paper provides an overview of recent techniques for improving non-adversarial robustness in deep learning for computer vision, summarizing methods, benchmark datasets, and evaluating their strengths and limitations.
Non-adversarial robustness, also known as natural robustness, is a property of deep learning models that enables them to maintain performance even when faced with distribution shifts caused by natural variations in data. However, achieving this property is challenging because it is difficult to predict in advance the types of distribution shifts that may occur. To address this challenge, researchers have proposed various approaches, some of which anticipate potential distribution shifts, while others utilize knowledge about the shifts that have already occurred to enhance model generalizability. In this paper, we present a brief overview of the most recent techniques for improving the robustness of computer vision methods, as well as a summary of commonly used robustness benchmark datasets for evaluating the model's performance under data distribution shifts. Finally, we examine the strengths and limitations of the approaches reviewed and identify general trends in deep learning robustness improvement for computer vision.