An extension of McDiarmid's inequality

2015/11/17 by Combes, Richard · 5 citations
#FOS: Computer and information sciences #FOS: Mathematics #Machine Learning (cs.LG) #Probability (math.PR) #Statistics Theory (math.ST)

paper · doi:10.48550/arxiv.1511.05240

Abstract

We generalize McDiarmid's inequality for functions with bounded differences on a high probability set, using an extension argument. Those functions concentrate around their conditional expectations. We further extend the results to concentration in general metric spaces.

Cited by

• Quasi-interpolation with random sampling centers
• Sampling Permutations with Cell Probes is Hard
• Rethinking Federated Learning Over the Air: The Blessing of Scaling Up
• Quantifying and testing dependence to categorical variables
• Convergence of empirical Gromov-Wasserstein distance

Related

• On McDiarmid's inequality for Hamming distance
• Concentration in unbounded metric spaces and algorithmic stability
• Distribution-dependent concentration inequalities for tighter generalization bounds
• Concentration inequality using unconfirmed knowledge
• On the application of McDiarmid's inequality to complex systems