$\mu$-tempered metadynamics: Artifact independent convergence times for wide hills
Bradley M. Dickson

TL;DR
This paper introduces $$-tempered metadynamics, a new scheme that achieves artifact-independent convergence times, enabling faster and more reliable free energy calculations with wide hills.
Contribution
The paper proposes $$-tempered metadynamics, a novel tempering method that improves convergence efficiency and reduces artifacts in free energy estimations.
Findings
Artifact-independent convergence times achieved with $$-tempered metadynamics.
Fast convergence demonstrated with wide hills up to 60b0.
Comparison shows improved efficiency over well-tempered metadynamics.
Abstract
Recent analysis of well-tempered metadynamics (WTmetaD) showed that it converges without mollification artifacts in the bias potential. Here we explore how metadynamics heals mollification artifacts, how healing impacts convergence time, and whether alternative temperings may be used to improve efficiency. We introduce "-tempered" metadynamics as a simple tempering scheme, inspired by a related mollified adaptive biasing potential (mABP), that results in artifact independent convergence of the free energy estimate. We use a toy model to examine the role of artifacts in WTmetaD and solvated alanine dipeptide to compare the well-tempered and -tempered frameworks demonstrating fast convergence for hill widths as large as for TmetaD.
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.
