A FISTA-based method with a custom PAVA computes perspective-relaxation dual bounds for k-sparse GLMs in O(p log p) per prox evaluation, enabling larger optimality certificates.
For this experiment, we vary the feature dimension p∈ {1000,2000,4000,8000,16000}
1 Pith paper cite this work. Polarity classification is still indexing.
1
Pith paper citing it
fields
cs.LG 1years
2025 1verdicts
CONDITIONAL 1representative citing papers
citing papers explorer
-
Scalable First-order Method for Certifying Optimal k-Sparse GLMs
A FISTA-based method with a custom PAVA computes perspective-relaxation dual bounds for k-sparse GLMs in O(p log p) per prox evaluation, enabling larger optimality certificates.