A finite-bank MNIST construction shows that an empirical estimator of algorithmic mutual information I(S;W) monotonically tracks the generalization gap in MoE models while providing an accuracy-rate curve via Blahut-Arimoto.
Local privacy and statistical minimax rates
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
years
2026 2representative citing papers
At typical differential privacy levels, Cox models lose significance for about 90% of covariates and drop to random predictive performance, with usable results requiring much weaker privacy.
citing papers explorer
-
Expert Routing for Communication-Efficient MoE via Finite Expert Banks
A finite-bank MNIST construction shows that an empirical estimator of algorithmic mutual information I(S;W) monotonically tracks the generalization gap in MoE models while providing an accuracy-rate curve via Blahut-Arimoto.
-
Benchmarking the Utility of Privacy-Preserving Cox Regression Under Data-Driven Clipping Bounds: A Multi-Dataset Simulation Study
At typical differential privacy levels, Cox models lose significance for about 90% of covariates and drop to random predictive performance, with usable results requiring much weaker privacy.