In: Journal of computational science, 2021, vol. 53, p. 13
The ℓ1-regularized Gaussian maximum likelihood method is a common approach for sparse precision matrix estimation, but one that poses a computational challenge for high-dimensional datasets. We present a novel ℓ1- regularized maximum likelihood method for performant large-scale sparse precision matrix estimation utilizing the block structures in the underlying computations. We identify the...
|
In: Biometrika, 2018, vol. 105, no. 3, p. 575-592
|
In: Statistical Applications in Genetics and Molecular Biology, 2017, vol. 16, no. 5-6, p. 291-312
|
In: Mathematical Geosciences, 2015, vol. 47, no. 7, p. 771-789
|
In: Theory and Decision, 2015, vol. 78, no. 2, p. 189-208
|
In: Geometriae Dedicata, 2015, vol. 177, no. 1, p. 367-384
|
In: Entropy, 2019, vol. 21, no. 8, p. 758
Variational inference with a factorized Gaussian posterior estimate is a widely-used approach for learning parameters and hidden variables. Empirically, a regularizing effect can be observed that is poorly understood. In this work, we show how mean field inference improves generalization by limiting mutual information between learned parameters and the data through noise. We quantify a...
|
In: Methodology and Computing in Applied Probability, 2014, vol. 16, no. 1, p. 169-185
|
In: Mathematische Zeitschrift, 2014, vol. 276, no. 3-4, p. 635-654
|
In: Methodology and Computing in Applied Probability, 2014, vol. 16, no. 2, p. 263-282
|