New method reduces distraction effects in image AI explanations
Seeing Through Distractions: Context-Stable Attribution via the Least Core
Computer Vision and Pattern RecognitionComputer Science and Game Theory
Summary
Sometimes, parts of a picture can seem unhelpful or distracting when AI tries to explain why it made a decision, even if those parts don't change the final answer. The authors studied these distracting parts, called contextual distractors, and found ways to better measure their importance by looking at different sized groups of features. They tested their ideas with experiments and found their method helps make explanations clearer and more reliable. This could help people understand AI decisions in images much better.
What this means in practice
- •For computer vision engineers: Improve explanations of image classification models by reducing misleading attributions from distracting features.
- •For machine learning interpretability teams: Create more stable and context-aware feature importance scores for visual AI systems using theoretical guarantees from least core methods.
Authors
Sai Ganesh Nagarajan, Toshinori Yamauchi, Hiroshi Kera
Abstract
Shapley-value-based explanations are widely used for feature attribution in machine learning. However, in computer vision, a feature may reduce the target-class score when added to the complete genuine image, even when the predicted label remains unchanged. We formalize such features as contextual distractors and investigate their attributions under semivalues and the least core. Under a weak average marginal monotonicity assumption, we derive quantitative semivalue--least-core gap bounds and sufficient conditions based on the probability of sampling smaller contexts. We further show that shifting semivalue weights toward larger contexts decreases the signed gap. Finally, controlled image-classification experiments and evaluations on Salient ImageNet provide strong empirical support for our theoretical findings.