Training Energy-Based Normalizing Flow with Score-Matching Objectives
Chao, Chen-Hao, Sun, Wei-Fang, Hsu, Yen-Chang, Kira, Zsolt, Lee, Chun-Yi
In this paper, we establish a connection between the parameterization of flow-based and energy-based generative models, and present a new flow-based modeling approach called energy-based normalizing flow (EBFlow). We demonstrate that by optimizing EBFlow with score-matching objectives, the computation of Jacobian determinants for linear transformations can be entirely bypassed. This feature enables the use of arbitrary linear layers in the construction of flow-based models without increasing the computational time complexity of each training iteration from $O(D^2L)$ to $O(D^3L)$ for an $L$-layered model that accepts $D$-dimensional inputs. This makes the training of EBFlow more efficient than the commonly-adopted maximum likelihood training method. In addition to the reduction in runtime, we enhance the training stability and empirical performance of EBFlow through a number of techniques developed based on our analysis of the score-matching methods. The experimental results demonstrate that our approach achieves a significant speedup compared to maximum likelihood estimation while outperforming prior methods with a noticeable margin in terms of negative log-likelihood (NLL).
Oct-28-2023
- Country:
- Asia > Taiwan (0.04)
- North America > United States
- Georgia > Fulton County
- Atlanta (0.04)
- California > Santa Clara County
- Santa Clara (0.04)
- Mountain View (0.04)
- Georgia > Fulton County
- Europe > United Kingdom
- England > Cambridgeshire > Cambridge (0.04)
- Genre:
- Research Report > New Finding (0.66)