Data Leverage References

Huaiguang Cai

Proposes CHG (compound of Hardness and Gradient) utility function to approximate the utility of each data subset, reducing computational complexity to a single model retraining—achieving a quadratic improvement over existing Data Shapley methods.

Junwei Deng, Weijing Tang, Jiaqi W. Ma

Proposes Versatile Influence Function (VIF) designed to fully leverage auto-differentiation, eliminating case-specific derivations. Demonstrated across Cox regression for survival analysis, node embedding for network analysis, and listwise learning-to-rank, with estimates closely resembling leave-one-out retraining while being up to 10^3 times faster.