“We had to control how big a number shows up as we do this guessing and coordination,” said Peng. Peng and Vempala prove that their algorithm can solve any sparse linear system in n2.332 steps. This ...

We propose an ℓ₁-penalized algorithm for fitting high-dimensional generalized linear mixed models (GLMMs). GLMMs can be viewed as an extension of generalized linear models for clustered observations.