Annealing in variational inference mitigates mode collapse: A theoretical study on Gaussian mixtures
stat.ML, cs.LG
Submitted: 2026-02-13
Updated: 2026-11-09
Comments: Accepted to NeurIPS 2026
Code: https://github.com/Luijios/annealed_vi_gmms
License: http://creativecommons.org/licenses/by/4.0/
The gist: Mode collapse, the failure to capture one or more modes when targeting a multimodal distribution, is a central challenge in modern variational inference.
Terminology
Abstract
Mode collapse, the failure to capture one or more modes when targeting a multimodal distribution, is a central challenge in modern variational inference. In this work, we provide a mathematical analysis of annealing based strategies for mitigating mode collapse in a tractable setting: learning a Gaussian mixture, where mode collapse is known to arise. Leveraging a low dimensional summary statistics description, we precisely characterize the interplay between the initial temperature and the annealing rate, and derive a sharp formula for the probability of mode collapse. Our analysis shows that an appropriately chosen annealing scheme can robustly prevent mode collapse. Finally, we present numerical evidence that these theoretical tradeoffs qualitatively extend to neural network based models, RealNVP normalizing flows, providing guidance for designing annealing strategies mitigating mode collapse in practical variational inference pipelines.
Sources
- Flow-based generative models for Markov chain Monte Carlo in lattice field theory
- Variational Neural Annealing
- Natural Variational Annealing for Multimodal Optimization
- Mitigating mode collapse in normalizing flows by annealing with an adaptive schedule: Application to parameter estimation
- Solving Statistical Mechanics Using Variational Autoregressive Networks
Related papers
- Uniform-in-Time Weak Propagation-of-Chaos in Shallow Neural Networks
- Diffusion Models Observe Only Gradients: A Geometric Perspective on Score Matching Errors
- Quantifying Error Propagation and Model Collapse in Diffusion Models
- Behavior of prediction performance metrics with rare events
- Statistical Inference and Learning for Shapley Additive Explanations (SHAP)
- Control Variate Design for Score-Function Gradient Estimators in Importance-Weighted Variational Inference