I am diving into some literature to understand which is the best algorithm for computing the log-determinant of a PSD matrix. More generally, I am interested in a list of resources to read, which cover all the possible cases of matrix (i.e. sparse, dense, low rank, etc..) with all the probability of failure and error (relative, absolute, etc..).
So far I have found the following two papers:
- Large-scale Log-determinant Computationthrough Stochastic Chebyshev Expansions
- Fast Estimation of $tr(f(A))$ via Stochastic Lanczos Quadrature
- A Randomized Algorithm for Approximating the Log Determinant of a Symmetric Positive Definite Matrix
We were just wondering if there are other algorithms with better asymptotics, different techniques, or other paper that we should be aware of.
Similar question on Computational Science are: