The Unified Non-Convex Framework for Robust Causal Inference: Overcoming the Gaussian Barrier and Optimization Fragility
This work addresses the challenge of reliable causal effect estimation in the presence of outliers and non-convexity, which is critical for fields like economics and healthcare, though it appears incremental as it builds on existing methods like gamma-Divergence and Graduated Non-Convexity.
The paper tackles the problem of robust causal inference by proposing a unified framework that overcomes Gaussian limitations and optimization fragility, achieving a 15% reduction in mean squared error on synthetic data and improved stability in high-dimensional settings.
This document proposes a Unified Robust Framework that re-engineers the estimation of the Average Treatment Effect on the Overlap (ATO). It synthesizes gamma-Divergence for outlier robustness, Graduated Non-Convexity (GNC) for global optimization, and a "Gatekeeper" mechanism to address the impossibility of higher-order orthogonality in Gaussian regimes.