Bridging Interpretability and Robustness Using LIME-Guided Model Refinement
It addresses critical vulnerabilities in deep learning for applications requiring reliable and interpretable AI, but it is incremental as it builds on existing interpretability methods.
This paper tackles the problem of deep learning models being vulnerable to adversarial attacks and lacking transparency by proposing a framework that uses LIME to refine models, resulting in improved robustness and interpretability on benchmark datasets.
This paper explores the intricate relationship between interpretability and robustness in deep learning models. Despite their remarkable performance across various tasks, deep learning models often exhibit critical vulnerabilities, including susceptibility to adversarial attacks, over-reliance on spurious correlations, and a lack of transparency in their decision-making processes. To address these limitations, we propose a novel framework that leverages Local Interpretable Model-Agnostic Explanations (LIME) to systematically enhance model robustness. By identifying and mitigating the influence of irrelevant or misleading features, our approach iteratively refines the model, penalizing reliance on these features during training. Empirical evaluations on multiple benchmark datasets demonstrate that LIME-guided refinement not only improves interpretability but also significantly enhances resistance to adversarial perturbations and generalization to out-of-distribution data.