[…] a faster way of computing an (approximate) solution to the least-squares problem (3). As we have seen before on this blog, sketching provides a natural approach to quickly and approximately solving a […]
[…] we need a faster way of computing an (approximate) solution to the least-squares problem (3). As we have seen before on this blog, sketching provides a natural approach to quickly and approximately […]
[…] faster way of computing an (approximate) solution to the least-squares problem (3). As we have seen before on this blog, sketching provides a natural approach to quickly and approximately solving a […]
Today, I want to talk about the generalized Nyström approximation, which I regard as the one of the “big three” approaches to constructing a low-rank approximation to matrix.1 Understanding this approximation, under what conditions it works…
[…] The vector inherits the isotropy property as well . As a consequence, we can use the vector to form an unbiased estimator for the matrix trace . Trace estimation has been a frequent topic on this blog. […]
Let be random vectors and let denote their tensor product. Assume the vectors are isotropic, in the sense that The vector inherits the isotropy property as well . As a consequence, we can use the vector to form an unbiased estimator for the…
I am a Miller research fellow in the mathematics department at UC Berkeley. My research is focused on designing computational techniques for solving large-scale problems in machine learning, quantum information, and scientific computing, wi…