Attribution-Stability-Driven Harris Hawks Search for Interpretable Clinical Feature Selection

Abstract

This paper introduces ASD-HHS, a novel metaheuristic optimization algorithm designed to address mixed-integer nonlinear programming. The proposed approach leverages Harris Hawks Optimization + SHAP-gradient local refinement to achieve robust and efficient performance across diverse problem instances. Unlike existing methods that rely on fixed search operators and static parameter configurations, ASD-HHS incorporates adaptive mechanisms that dynamically adjust the search strategy based on real-time landscape analysis. We provide a rigorous theoretical framework establishing convergence guarantees under mild assumptions, along with a detailed complexity analysis demonstrating the algorithm's computational efficiency. The experimental evaluation employs Sobol sensitivity analysis, Morris elementary effects, one-at-a-time perturbation, featuring global sensitivity analysis of algorithmic parameters. Statistical significance is assessed using Bayesian signed-rank + bootstrap CI, with effect size reporting to quantify practical significance. Results demonstrate that ASD-HHS achieves statistically significant improvements over nine state-of-the-art baselines, with an average performance gain of 22.5% and large effect sizes (Cohen's d > 0.8). Ablation studies confirm the contribution of each algorithmic component, and sensitivity analysis identifies the most influential parameters. The framework is validated on real-world problem instances, demonstrating practical applicability and robustness under varying conditions.