A novel framework for free energy estimation that leverages learned transports implemented via stochastic interpolants and provides consistent, minimum-variance estimators based on escorted Jarzynski equality and controlled Crooks theorem, alongside variational upper and lower bounds on free energy differences.
Jiajun He, Yuan-Qi Du, Francisco Vargas et al.· Neural Information Processin...· 23 citations· ⚡1
This work shows that a direction-dependent schedule decomposes into two independent design choices: a variance path, which fully determines the intermediate laws and probability flow, and a factorization, which leaves this flow unchanged while controlling irreducible regression variance.
A. Claustre, Hugo Negrel, Claire Boyer et al.· 1 citation
An instantiation of flow matching that relies on a time-independent velocity field to exactly map between two distributions, so long as the target is singular, is proposed and supported on a lower-dimensional data manifold.
Cheuk-Kit Lee, Florentin Coeurdoux, Yuyuan Chen et al.· 4 citations
Generating samples from limited information is a fundamental problem across scientific domains. Classical maximum entropy methods provide principled uncertainty quantification from moment constraints but require sampling via MCMC or Langevin dynamics, which typically exhibit exponential slowdown in high dimensions. In...
