Reasoning with trees: interpreting CNNs using hierarchies
Abstract: Challenges remain in providing interpretable explanations for neural network reasoning in explainable AI (xAI). Existing methods like Integrated Gradients produce noisy maps, and LIME, while intuitive, may deviate from the model's reasoning. We introduce a framework that uses hierarchical segmentation techniques for faithful and interpretable explanations of Convolutional Neural Networks (CNNs). Our method constructs model-based hierarchical segmentations that maintain the model's reasoning fidelity and allow both human-centric and model-centric segmentation. This approach can be combined with various xAI methods and provides multiscale explanations that help identify biases and improve understanding of neural network decision-making. Experiments show that our framework, xAiTrees, delivers highly interpretable and faithful model explanations, not only surpassing traditional xAI methods but shedding new light on a novel approach to enhancing xAI interpretability.
Submission Length: Regular submission (no more than 12 pages of main content)
Assigned Action Editor: ~Karthikeyan_Shanmugam1
Submission Number: 4440
Loading