On the Non-robustness of Essentially Conditional Information Inequalities

Tarik Kaced 1, * Andrei Romashchenko 1, *
* Corresponding author
1 ESCAPE - Systèmes complexes, automates et pavages
LIRMM - Laboratoire d'Informatique de Robotique et de Microélectronique de Montpellier
Abstract : We show that two essentially conditional linear inequalities for Shannon's entropies (including the Zhang- Yeung'97 conditional inequality) do not hold for asymptotically entropic points. This means that these inequalities are non-robust in a very strong sense. This result raises the question of the meaning of these inequalities and the validity of their use in practice-oriented applications.
Complete list of metadatas

Cited literature [20 references]  Display  Hide  Download

https://hal-lirmm.ccsd.cnrs.fr/lirmm-00736192
Contributor : Tarik Kaced <>
Submitted on : Thursday, September 27, 2012 - 5:12:20 PM
Last modification on : Thursday, May 24, 2018 - 3:59:23 PM
Long-term archiving on: Friday, December 28, 2012 - 8:30:08 AM

Identifiers

  • HAL Id : lirmm-00736192, version 1

Collections

Citation

Tarik Kaced, Andrei Romashchenko. On the Non-robustness of Essentially Conditional Information Inequalities. ITW'12: Information Theory Workshop, Sep 2012, Switzerland. pp.1935-1939. ⟨lirmm-00736192⟩

Share

Metrics

Record views

344

Files downloads

470