On Measuring Intrinsic Causal Attributions in Deep Neural Networks
By: Saptarshi Saha , Dhruv Vansraj Rathore , Soumadeep Saha and more
Potential Business Impact:
Shows how computer brains make decisions.
Quantifying the causal influence of input features within neural networks has become a topic of increasing interest. Existing approaches typically assess direct, indirect, and total causal effects. This work treats NNs as structural causal models (SCMs) and extends our focus to include intrinsic causal contributions (ICC). We propose an identifiable generative post-hoc framework for quantifying ICC. We also draw a relationship between ICC and Sobol' indices. Our experiments on synthetic and real-world datasets demonstrate that ICC generates more intuitive and reliable explanations compared to existing global explanation techniques.
Similar Papers
Causality-Driven Neural Network Repair: Challenges and Opportunities
Machine Learning (CS)
Fixes AI mistakes by understanding why they happen.
A Causal Framework for Aligning Image Quality Metrics and Deep Neural Network Robustness
CV and Pattern Recognition
Improves AI's understanding of image quality.
Attribution Explanations for Deep Neural Networks: A Theoretical Perspective
Machine Learning (CS)
Makes AI decisions easier to understand.