Optimizing Computational-Statistical Runtime for Wasserstein Distance Estimation
Researchers have developed a new method to optimize the computational-statistical runtime for estimating Wasserstein distance. This technique, called Sample-Sketch-Solve, uses a regular cartesian grid to sketch data, which compresses it without increasing asymptotic error. The approach enables faster exact algorithms and approximates the Wasserstein-2 squared distance within epsilon error in a time complexity that is optimal for certain smooth distributions. AI
IMPACT Improves efficiency for a core statistical tool used in machine learning model evaluation.