Optimizing NOTEARS Objectives via Topological Swaps

Abstract

Recently, an intriguing class of non-convex optimization problems has emerged in the context of learning directed acyclic graphs (DAGs). These problems involve minimizing a given loss or score function, subject to a non-convex continuous constraint that penalizes the presence of cycles in a graph. In this work, we delve into the optimality challenges associated with this class of non-convex programs. To address these challenges, we propose a bi-level algorithm that leverages the non-convex constraint in a novel way. The outer level of the algorithm optimizes over topological orders by iteratively swapping pairs of nodes within the topological order of a DAG. A key innovation of our approach is the development of an effective method for generating a set of candidate swapping pairs for each iteration. At the inner level, given a topological order, we utilize off-the-shelf solvers that can handle linear constraints. The key advantage of our proposed algorithm is that it is guaranteed to find a local minimum or a KKT point under weaker conditions compared to previous work and finds solutions with lower scores. Extensive experiments demonstrate that our method outperforms state-of-the-art approaches in terms of achieving a better score. Additionally, our method can also be used as a post-processing algorithm to significantly improve the score of other algorithms. Code implementing the proposed method is available at https://github.com/duntrain/topo.

Cite

Text

Deng et al. "Optimizing NOTEARS Objectives via Topological Swaps." International Conference on Machine Learning, 2023.

Markdown

[Deng et al. "Optimizing NOTEARS Objectives via Topological Swaps." International Conference on Machine Learning, 2023.](https://mlanthology.org/icml/2023/deng2023icml-optimizing/)

BibTeX

@inproceedings{deng2023icml-optimizing,
  title     = {{Optimizing NOTEARS Objectives via Topological Swaps}},
  author    = {Deng, Chang and Bello, Kevin and Aragam, Bryon and Ravikumar, Pradeep Kumar},
  booktitle = {International Conference on Machine Learning},
  year      = {2023},
  pages     = {7563-7595},
  volume    = {202},
  url       = {https://mlanthology.org/icml/2023/deng2023icml-optimizing/}
}