2021/09/23 by A. Emin Orhan, Orhan, A. Emin
Computer Science · Engineering · #Computer Vision and Pattern Recognition (cs.CV) #Domain Adaptation and Few-Shot Learning #FOS: Computer and information sciences #Image Processing Techniques and Applications #Machine Learning (cs.LG) #Neural and Evolutionary Computing (cs.NE) #Remote-Sensing Image Classification
paper · pdf · doi:10.48550/arxiv.2109.11523
openalex publication_date 2021/09/23 · openalex created_date 2022/07/25 · openalex updated_date 2026/07/28
This paper addresses a fundamental question: how good are our current self-supervised visual representation learning algorithms relative to humans? More concretely, how much "human-like" natural visual experience would these algorithms need in order to reach human-level performance in a complex, realistic visual object recognition task such as ImageNet? Using a scaling experiment, here we estimate that the answer is several orders of magnitude longer than a human lifetime: typically on the order of a million to a billion years of natural visual experience (depending on the algorithm used). We obtain even larger estimates for achieving human-level performance in ImageNet-derived robustness benchmarks. The exact values of these estimates are sensitive to some underlying assumptions, however even in the most optimistic scenarios they remain orders of magnitude larger than a human lifetime. We discuss the main caveats surrounding our estimates and the implications of these surprising results.