Year: 2019
Type: article
Abstract: Does the human mind resemble the machine-learning systems that mirror its performance? Convolutional neural networks (CNNs) have achieved human-level benchmarks in classifying novel images. These advances support technologies such as autonomous vehicles and machine diagnosis; but beyond this, they serve as candidate models for human vision itself. However, unlike humans, CNNs are "fooled" by adversarial examples-nonsense patterns that machines recognize as familiar objects, or seemingly irrelevant image perturbations that nevertheless alter the machine's classification. Such bizarre behaviors challenge the promise of these new advances; but do human and machine judgments fundamentally diverge? Here, we show that human and machine classification of adversarial images are robustly related: In 8 experiments on 5 prominent and diverse adversarial imagesets, human subjects correctly anticipated the machine's preferred label over relevant foils-even for images described as "totally unrecognizable to human eyes". Human intuition may be a surprisingly reliable guide to machine (mis)classification-with consequences for minds and machines alike.
Source: Nature Communications
Authors: Zhenglong Zhou, Chaz Firestone
Institution: Johns Hopkins University
Language: English
Subfield: Artificial Intelligence
Field: Computer Science
Domain: Physical Sciences
Open Access status: gold
Estimated APC paid: $5,200
OpenAlex ID: W2919093153