On the accuracy of the normal approximation for the free energy in the REM
Raphael Meiners, Anselm Reichenbachs

TL;DR
This paper investigates the accuracy of the normal approximation for free energy fluctuations in the random energy model, revealing its limitations at high temperatures and specific scalings.
Contribution
It provides a detailed analysis of the regimes where the normal approximation is valid and identifies the decay rate of moderate deviations.
Findings
Normal approximation holds only in a narrow scaling range at high temperatures.
Probabilities of moderate deviations decay faster than exponentially for higher order scalings.
The results clarify the limitations of CLT-based approximations in the REM.
Abstract
In the present paper we consider the fluctuations of the free energy in the random energy model (REM) on a moderate deviation scale. We find that for high temperatures the normal approximation holds only in a narrow range of scalings away from the CLT. For scalings of higher order, probabilities of moderate deviations decay faster than exponentially.
Peer Reviews
No public reviews on file for this paper yet. If you reviewed it on a platform where reviews are public (OpenReview, ICLR, NeurIPS, ICML), you can paste yours below so the community can read it here.
Videos
No videos yet. Explain this paper in a talk, walkthrough, or lecture? Add one.
