Opti-CAM: Optimizing saliency maps for interpretability

2023/01/17 by Zhang, Hanwei, Torres, Felipe, Sicre, Ronan +2 · 4 citations
#Computer Vision and Pattern Recognition (cs.CV) #FOS: Computer and information sciences

paper · doi:10.48550/arxiv.2301.07002

Abstract

Methods based on class activation maps (CAM) provide a simple mechanism to interpret predictions of convolutional neural networks by using linear combinations of feature maps as saliency maps. By contrast, masking-based methods optimize a saliency map directly in the image space or learn it by training another network on additional data. In this work we introduce Opti-CAM, combining ideas from CAM-based and masking-based approaches. Our saliency map is a linear combination of feature maps, where weights are optimized per image such that the logit of the masked image for a given class is maximized. We also fix a fundamental flaw in two of the most common evaluation metrics of attribution methods. On several datasets, Opti-CAM largely outperforms other CAM-based approaches according to the most relevant classification metrics. We provide empirical evidence supporting that localization and classifier interpretability are not necessarily aligned.

Cited by

• Back to the Baseline: Examining Baseline Effects on Explainability Metrics
• CaberNet: Causal Representation Learning for Cross-Domain HVAC Energy Prediction
• There is More to Attention: Statistical Filtering Enhances Explanations in Vision Transformers
• Foveated Retinotopy Improves Classification and Localization in Convolutional Neural Networks

Related

• Poly-CAM: High resolution class activation map for convolutional neural networks
• Rethinking gradient weights' influence over saliency map estimation
• Investigating and Simplifying Masking-based Saliency Methods for Model\n Interpretability
• Abs-CAM: A Gradient Optimization Interpretable Approach for Explanation of Convolutional Neural Networks
• SCAAT: Improving Neural Network Interpretability via Saliency Constrained Adaptive Adversarial Training