activity
20112023
most citedLearning Representations for Neural Network-Based Classification Using the Information Bottleneck Principle

200 citations · 675 across the 45 of their papers we have counts for

collaborators
Showing math.PRShow all

2 papers · 1 filter

math.PR2016★ 4 cited

A Sufficient Condition for a Unique Invariant Distribution of a Higher-Order Markov Chain

Bernhard C. Geiger

We derive a sufficient condition for a -th order homogeneous Markov chain with finite alphabet to have a unique invariant distribution on $\mathcal{Z}…

math.PR2015

Cepstral Analysis of Random Variables: Muculants

Christian Knoll, Bernhard C. Geiger, Gernot Kubin

An alternative parametric description for discrete random variables, called muculants, is proposed. In contrast to cumulants, muculants are based on the Fourier series expansion, r…