Diffusion Methods for Generating Transition Paths
This work addresses the challenge of generating high-quality transition paths for molecular systems where data is often scarce, providing an efficient method for studying rare transitions.
The paper tackled the problem of simulating rare transitions between metastable states in molecular systems by developing two novel score-based generative methods, a chain-based and a midpoint-based approach, which demonstrated effectiveness in generating transition paths for the Müller potential and Alanine dipeptide in both data-rich and data-scarce regimes.
In this work, we seek to simulate rare transitions between metastable states using score-based generative models. An efficient method for generating high-quality transition paths is valuable for the study of molecular systems since data is often difficult to obtain. We develop two novel methods for path generation in this paper: a chain-based approach and a midpoint-based approach. The first biases the original dynamics to facilitate transitions, while the second mirrors splitting techniques and breaks down the original transition into smaller transitions. Numerical results of generated transition paths for the Müller potential and for Alanine dipeptide demonstrate the effectiveness of these approaches in both the data-rich and data-scarce regimes.