Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Watch a 55-minute lecture by Tudor Manole from MIT presenting "Sharp Deconvolution of Optimal Transport Matchings" at IPAM's Statistical and Numerical Methods for Non-commutative Optimal Transport Workshop at UCLA. Recorded on May 20, 2025, this talk explores the statistical challenge of recovering optimal transport matching between discrete measures from noisy convolutions with smooth kernels. Learn about the minimax risk characterization for estimating discrete measures under Wasserstein distance, with applications in super-resolution microscopy where optimal transport matchings help quantify spatial proximity between discrete signals. Discover how simple estimators based on maximum likelihood estimation achieve these minimax rates. The presentation covers joint research with Shayan Hundrieser, Danila Litskevich, and Axel Munk.
