Adaptive Independence Test via the Generalized HSIC

Learn about adaptive independence testing methods through the generalized Hilbert-Schmidt Independence Criterion (HSIC) in this 55-minute conference talk from BIMSA's ICBS2025 series. Explore advanced statistical techniques for testing independence between random variables using kernel-based methods, with focus on adaptive approaches that can automatically adjust to different data characteristics and distributions. Discover how the generalized HSIC framework extends traditional independence testing capabilities and examine practical applications in statistical analysis and machine learning contexts where determining variable independence is crucial for model validity and interpretation.