Accelerating Continuous Normalizing Flow with Trajectory Polynomial Regularization

Understand

In this paper, we propose an approach to effectively accelerating the computation of continuous normalizing flow (CNF), which has been proven to be a powerful tool for the tasks such as variational inference and density estimation.

• The training time cost of CNF can be extremely high because the required number of function evaluations (NFE) for solving corresponding ordinary differential equations (ODE) is very large.
• We think that the high NFE results from large truncation errors of solving ODEs.
• To address the problem, we propose to add a regularization.