PreprintarXiv licencearXiv
On counterexamples to the unification of free entropy
Abstract
It has been a long-standing open problem to determine when two versions of Voiculescu's free entropy, χ and χ*, agree. OpenAI recently generated the first example with −∞ < χ< χ*. We give a detailed presentation of a similar example, and explicitly show the separation between free Fisher information and the large-n limit of classical Fisher information through the relationship with Wasserstein distance. We thus situate the example in the context of probability theory and information geometry.
Related
- Counterexamples to additivity of minimum output p-Rényi entropy of quantum channels for all p≥ 0
- Near-Gaussian counterexamples to the Ball-Nayar-Tkocz entropy concavity conjecture
- Counterexamples to two conjectures on the diameter of clique-free graphs
- Entropy Regularization: A Free Correction to Cross-Entropy for Verified Demonstrations
- Entropy-Regularized Mean-Field Games with Average-Reward Criterion: Span-Seminorm Contractions and Model-Free Learning
BibTeX
@misc{jekel2026counterexamples,
title = {{On counterexamples to the unification of free entropy}},
author = {David Jekel},
howpublished = {arXiv},
year = {2026},
month = oct,
doi = {10.48550/arxiv.2610.27243},
eprint = {2610.27243},
archivePrefix = {arXiv},
url = {https://arxiv.org/abs/2610.27243},
}
RIS
TY - UNPB
TI - On counterexamples to the unification of free entropy
AU - David Jekel
PY - 2026
DA - 2026/10/19
PB - arXiv
DO - 10.48550/arxiv.2610.27243
UR - https://arxiv.org/abs/2610.27243
AB - It has been a long-standing open problem to determine when two versions of Voiculescu's free entropy, χ and χ^*, agree. OpenAI recently generated the first example with −∞ < χ< χ^*. We give a detailed presentation of a similar example, and explicitly show the separation between free Fisher information and the large-n limit of classical Fisher information through the relationship with Wasserstein distance. We thus situate the example in the context of probability theory and information geometry.
ER -