Coursera Flash Sale
40% Off Coursera Plus for 3 Months!
Grab it
Explore a comprehensive toolkit for implicit diagonal estimation in this 11-minute JuliaCon 2024 talk by Niclas Popp. Dive into RandomizedDiagonalEstimation.jl, which offers established algorithms for approximating the diagonal of large matrices using only matrix-vector products. Learn about applications in network science, material science, and machine learning. Discover the Girard-Hutchinson estimator and Diag++ method, as well as novel extensions for estimating the diagonal of matrix functions like exponentials and inverses. Understand how the package integrates with the Julia ecosystem and allows for easy addition of future algorithms. Gain insights into this powerful tool for tackling problems in various scientific fields through efficient diagonal estimation techniques.
