A robustified vision model's ground-truth logit predicts human image difficulty, and logit-maximizing enhancements used in an easy-to-hard curriculum improve human visual category learning by 33-72% in margin above chance and cut training time by about 20%.
RN50 = ResNet-50, and ϵ values in the labels for each bar show the magnitude of the adversarial perturbations during adversarial training
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.CV 1years
2024 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
L-WISE: Boosting Human Visual Category Learning Through Model-Based Image Selection and Enhancement
A robustified vision model's ground-truth logit predicts human image difficulty, and logit-maximizing enhancements used in an easy-to-hard curriculum improve human visual category learning by 33-72% in margin above chance and cut training time by about 20%.