DAGs with No Curl: An Efficient DAG Structure Learning Approach

06/14/2021
by   Yue Yu, et al.
0

Recently directed acyclic graph (DAG) structure learning is formulated as a constrained continuous optimization problem with continuous acyclicity constraints and was solved iteratively through subproblem optimization. To further improve efficiency, we propose a novel learning framework to model and learn the weighted adjacency matrices in the DAG space directly. Specifically, we first show that the set of weighted adjacency matrices of DAGs are equivalent to the set of weighted gradients of graph potential functions, and one may perform structure learning by searching in this equivalent set of DAGs. To instantiate this idea, we propose a new algorithm, DAG-NoCurl, which solves the optimization problem efficiently with a two-step procedure: 1) first we find an initial cyclic solution to the optimization problem, and 2) then we employ the Hodge decomposition of graphs and learn an acyclic graph by projecting the cyclic graph to the gradient of a potential function. Experimental studies on benchmark datasets demonstrate that our method provides comparable accuracy but better efficiency than baseline DAG structure learning methods on both linear and generalized structural equation models, often by more than one order of magnitude.

READ FULL TEXT

page 1

page 2

page 3

page 4

research
09/16/2022

DAGMA: Learning DAGs via M-matrices and a Log-Determinant Acyclicity Characterization

The combinatorial problem of learning directed acyclic graphs (DAGs) fro...
research
09/15/2023

Constraint-Free Structure Learning with Smooth Acyclic Orientations

The structure learning problem consists of fitting data generated by a D...
research
08/28/2016

A Generalization of the Directed Graph Layering Problem

The Directed Layering Problem (DLP) solves a step of the widely used lay...
research
02/04/2023

Directed Acyclic Graphs With Tears

Bayesian network is a frequently-used method for fault detection and dia...
research
07/01/2021

Learning Large DAGs by Combining Continuous Optimization and Feedback Arc Set Heuristics

Bayesian networks represent relations between variables using a directed...
research
06/10/2020

Low Rank Directed Acyclic Graphs and Causal Structure Learning

Despite several important advances in recent years, learning causal stru...
research
03/04/2018

DAGs with NO TEARS: Smooth Optimization for Structure Learning

Estimating the structure of directed acyclic graphs (DAGs, also known as...

Please sign up or login with your details

Forgot password? Click here to reset