Skip to Main content Skip to Navigation
New interface
Conference papers

Regularized Frank-Wolfe for Dense CRFs: Generalizing Mean Field and Beyond

Ð.Khuê Lê-Huu 1 Karteek Alahari 1 
1 Thoth - Apprentissage de modèles à partir de données massives
Inria Grenoble - Rhône-Alpes, LJK - Laboratoire Jean Kuntzmann
Abstract : We introduce regularized Frank-Wolfe, a general and effective algorithm for inference and learning of dense conditional random fields (CRFs). The algorithm optimizes a nonconvex continuous relaxation of the CRF inference problem using vanilla Frank-Wolfe with approximate updates, which are equivalent to minimizing a regularized energy function. Our proposed method is a generalization of existing algorithms such as mean field or concave-convex procedure. This perspective not only offers a unified analysis of these algorithms, but also allows an easy way of exploring different variants that potentially yield better performance. We illustrate this in our empirical results on standard semantic segmentation datasets, where several instantiations of our regularized Frank-Wolfe outperform mean field inference, both as a standalone component and as an end-to-end trainable layer in a neural network. We also show that dense CRFs, coupled with our new algorithms, produce significant improvements over strong CNN baselines.
Complete list of metadata
Contributor : Đ.Khuê Lê-Huu Connect in order to contact the contributor
Submitted on : Wednesday, October 27, 2021 - 4:25:29 PM
Last modification on : Tuesday, February 8, 2022 - 12:49:29 PM
Long-term archiving on: : Friday, January 28, 2022 - 7:20:41 PM


Files produced by the author(s)


  • HAL Id : hal-03406107, version 1



Ð.Khuê Lê-Huu, Karteek Alahari. Regularized Frank-Wolfe for Dense CRFs: Generalizing Mean Field and Beyond. NeurIPS 2021 - 35th Annual Conference on Neural Information Processing Systems, Dec 2021, Virtual-only Conference, Australia. pp.1-35. ⟨hal-03406107⟩



Record views


Files downloads