Nash integrates neural networks into variational empirical Bayes to learn per-covariate penalties for sparse high-dimensional regression, claiming major speedups and better accuracy on real data.
A.1 Update for qβj Note that given that when qb, g(.; ., θ) are fixed then the objective for qβj j = 1
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
stat.ML 1years
2025 1verdicts
UNVERDICTED 1representative citing papers
citing papers explorer
-
Nash: Neural Adaptive Shrinkage for Structured High-Dimensional Regression
Nash integrates neural networks into variational empirical Bayes to learn per-covariate penalties for sparse high-dimensional regression, claiming major speedups and better accuracy on real data.