A nonconvex l1/2-regularized nonnegative matrix factorization method with ADMM solver and detection estimation improves sparse network recovery under imperfect observations compared to baselines.
Robust principal component analysis?
3 Pith papers cite this work. Polarity classification is still indexing.
3
Pith papers citing it
years
2026 3verdicts
UNVERDICTED 3representative citing papers
Under logarithmic loss, PCA on heavy-tailed observations from the superstatistical model recovers the principal directions of the underlying Gaussian generator's covariance.
SLaB compresses LLM weights via sparse-lowrank-binary decomposition guided by activation-aware scores, achieving up to 36% lower perplexity than prior methods at 50% compression on Llama models.
citing papers explorer
-
Sparse Network Inference under Imperfect Detection and its Application to Ecological Networks
A nonconvex l1/2-regularized nonnegative matrix factorization method with ADMM solver and detection estimation improves sparse network recovery under imperfect observations compared to baselines.
-
Heavy-Tailed Principal Component Analysis
Under logarithmic loss, PCA on heavy-tailed observations from the superstatistical model recovers the principal directions of the underlying Gaussian generator's covariance.
-
SLaB: Sparse-Lowrank-Binary Decomposition for Efficient Large Language Models
SLaB compresses LLM weights via sparse-lowrank-binary decomposition guided by activation-aware scores, achieving up to 36% lower perplexity than prior methods at 50% compression on Llama models.