Spline Approximation

Spline approximation is a powerful technique for representing complex functions and data using piecewise polynomial segments, aiming to balance accuracy and computational efficiency. Current research focuses on developing and applying spline-based models in diverse fields, including trajectory inference (using B-splines and Wasserstein space), generative modeling (via optimal transport and projection pursuit), and predictive analytics (with fair multivariate adaptive regression splines). These advancements improve the accuracy and interpretability of models across various applications, from fraud detection and cell biology to sound field estimation and physical system simulation.