Learning Diagonal Gaussian Mixture Models and Incomplete Tensor Decompositions

Explore a 29-minute conference talk on learning diagonal Gaussian mixture models and incomplete tensor decompositions. Delve into the mathematical foundations and algorithms for tensor computations, presented by Jiawang Nie from the University of California, San Diego. Discover how generating polynomials are utilized to compute incomplete symmetric tensor decompositions and approximations, and learn how these methods are applied to diagonal Gaussian mixture models. Examine the stability analysis, which demonstrates the high accuracy of obtained parameters when first and third order moments are sufficiently precise. Gain insights from this joint work with Bingni Guo and Zi Yang, presented at the Institute for Pure and Applied Mathematics' 2021 workshop on Tensor Methods and Emerging Applications to the Physical and Data Sciences.

Jiawang Nie: "Learning Diagonal Gaussian Mixture Models and Incomplete Tensor Decompositions"