Papers
Topics
Authors
Recent
Assistant
AI Research Assistant
Well-researched responses based on relevant abstracts and paper content.
Custom Instructions Pro
Preferences or requirements that you'd like Emergent Mind to consider when generating responses.
Gemini 2.5 Flash
Gemini 2.5 Flash 152 tok/s
Gemini 2.5 Pro 25 tok/s Pro
GPT-5 Medium 20 tok/s Pro
GPT-5 High 30 tok/s Pro
GPT-4o 92 tok/s Pro
Kimi K2 134 tok/s Pro
GPT OSS 120B 437 tok/s Pro
Claude Sonnet 4.5 37 tok/s Pro
2000 character limit reached

Entropic curvature not comparable to other curvatures -- or is it? (2404.04581v3)

Published 6 Apr 2024 in math.DG, math.MG, and math.PR

Abstract: In this paper we consider global $\theta$-curvatures of finite Markov chains with associated means $\theta$ in the spirit of the entropic curvature (based on the logarithmic mean) by Erbar-Maas and Mielke. As in the case of Bakry-\'Emery curvature, we also allow for a finite dimension parameter by making use of an adapted $\Gamma$ calculus for $\theta$-curvatures. We prove explicit positive lower curvature bounds (both finite- and infinite-dimensional) for finite abelian Cayley graphs. In the case of cycles, we provide also an upper curvature bound which shows that our lower bounds are asymptotically sharp (up to a logarithmic factor). Moreover, we prove new universal lower curvature bounds for finite Markov chains as well as curvature perturbation results (allowing, in particular, to compare entropic and Bakry-\'Emery curvatures). Finally, we present examples where entropic curvature differs significantly from other curvature notions like Bakry-\'Emery curvature or Ollivier Ricci and sectional curvatures.

Summary

We haven't generated a summary for this paper yet.

Dice Question Streamline Icon: https://streamlinehq.com

Open Questions

We haven't generated a list of open questions mentioned in this paper yet.

Lightbulb Streamline Icon: https://streamlinehq.com

Continue Learning

We haven't generated follow-up questions for this paper yet.

List To Do Tasks Checklist Streamline Icon: https://streamlinehq.com

Collections

Sign up for free to add this paper to one or more collections.