stat.MLFeb 13, 2026

Annealing in variational inference mitigates mode collapse: A theoretical study on Gaussian mixtures

Authors: Luigi Fogliani, Bruno Loureiro, Marylou Gabrié

Organizations: Departement d’Informatique, École Normale Supérieure, Université PSL, CNRS · Laboratoire de Physique de l’École Normale Supérieure, Université PSL, CNRS

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.

Figures & tables

Appendix figures & tables5 assets

Supplementary material from the paper’s appendix.

Appendix

Explore similar work

Oct 20, 2025

Mode Collapse of Mean-Field Variational Inference

Mean-field variational inference (MFVI) is a widely used method for approximating high-dimensional probability distributions by product measures. It has been empirically observed that MFVI optimizers often suffer from mode collapse. For example, when the target measure ππ is a mixture π=wP0+(1−w)P1π= w P_0 + (1 - w) P_1, the MFVI optimizer tends to place most of its mass near a single component of the mixture. This work provides the first theoretical explanation of mode collapse in MFVI. Starting from two-component mixtures, we introduce the notion of \emph{ε\varepsilon-separateness} to quantify the separation of the mixture components. We derive explicit bounds on the mass that any MFVI optimizer assigns to each component, when P0P_0 and P1P_1 are ε\varepsilon-separated for sufficiently small ε\varepsilon. We further extend this analysis to mixture target distributions with multiple components. Our results reveal that the occurrence of mode collapse crucially depends on the relative position of the components. To address this issue, we propose rotational variational inference (RoVI), which augments MFVI with a rotation matrix. The numerical studies support our theoretical findings and illustrate improved mode coverage with RoVI in Bayesian mixture posteriors, including a real-data example.
Nov 3, 2026

Transport along Local Annealing Paths: Wasserstein Geometry and Multimodal Sampling

Variational approaches to learning transport maps for multimodal sampling can suffer from mode collapse. One way to mitigate this problem is to learn transport along a prescribed path of intermediate distributions defined by geometric annealing. We show that annealing can take a substantial detour in Wasserstein space: when mass moves between modes, the path can be much longer than the distance between its endpoints. We further link path geometry to transport learning, showing that long annealing paths can slow population optimization, while sufficiently regular paths permit rapid convergence. Motivated by these findings, we propose componentwise annealed transport (CAT) to reduce the need for mass transfer between modes. Starting from a mode-informed mixture reference, CAT constructs an exact mixture decomposition of the target and learns separate transport maps along componentwise annealing paths. We estimate the component weights separately and show that these estimates remain consistent even when the learned flows are inexact. We also develop adaptive annealing schedules guided by transport effort. Experiments on multimodal synthetic distributions and Bayesian inference problems demonstrate improvements in transport learning and sampling efficiency.
Oct 2, 2022

GFlowNets and variational inference

This paper builds bridges between two families of probabilistic algorithms: (hierarchical) variational inference (VI), which is typically used to model distributions over continuous spaces, and generative flow networks (GFlowNets), which have been used for distributions over discrete structures such as graphs. We demonstrate that, in certain cases, VI algorithms are equivalent to special cases of GFlowNets in the sense of equality of expected gradients of their learning objectives. We then point out the differences between the two families and show how these differences emerge experimentally. Notably, GFlowNets, which borrow ideas from reinforcement learning, are more amenable than VI to off-policy training without the cost of high gradient variance induced by importance sampling. We argue that this property of GFlowNets can provide advantages for capturing diversity in multimodal target distributions.