• Corpus ID: 242757315

# Multi-task Learning of Order-Consistent Causal Graphs

@article{Chen2021MultitaskLO,
title={Multi-task Learning of Order-Consistent Causal Graphs},
author={Xinshi Chen and Haoran Sun and Caleb Ellington and Eric P. Xing and Le Song},
journal={ArXiv},
year={2021},
volume={abs/2111.02545}
}
• Published 3 November 2021
• Computer Science, Mathematics
• ArXiv
We consider the problem of discovering K related Gaussian directed acyclic graphs (DAGs), where the involved graph structures share a consistent causal order and sparse unions of supports. Under the multi-task learning setting, we propose a l 1 /l 2 regularized maximum likelihood estimator (MLE) for learning K linear structural equation models. We theoretically show that the joint estimator, by leveraging data across related tasks, can achieve a better sample complexity for recovering the…
1 Citations

## Figures and Tables from this paper

### On the Convergence of Continuous Constrained Optimization for Structure Learning

• Computer Science
AISTATS
• 2022
This work reviews the standard convergence result of the ALM and shows that the required conditions are not satisfied in the recent continuous constrained formulation for learning DAGs, and establishes the convergence guarantee of QPM to a DAG solution, under mild conditions, based on a property of the DAG constraint term.

## References

SHOWING 1-10 OF 57 REFERENCES

### High-dimensional joint estimation of multiple directed Gaussian graphical models

• Computer Science
Electronic Journal of Statistics
• 2020
It is proved that under certain regularity conditions, the proposed $\ell_0$-penalized maximum likelihood estimator converges in Frobenius norm to the adjacency matrices consistent with the data-generating distributions and has the correct sparsity.

### Integer Programming for Learning Directed Acyclic Graphs from Continuous Data

• Computer Science
ArXiv
• 2019
Learning directed acyclic graphs (DAGs) from data is a challenging task both in theory and in practice, because the number of possible DAGs scales superexponentially with the number of nodes. In this

### DAGs with NO TEARS: Continuous Optimization for Structure Learning

• Computer Science
NeurIPS
• 2018
This paper forms the structure learning problem as a purely continuous optimization problem over real matrices that avoids this combinatorial constraint entirely and achieves a novel characterization of acyclicity that is not only smooth but also exact.

### Inferring large graphs using l1-penalized likelihood

• Computer Science
• 2015
A novel procedure based on a specific formulation of the l1-norm regularized maximum likelihood is proposed, which decomposes the graph estimation into two optimization sub-problems: topological structure and node order learning.

### Learning Sparse Nonparametric DAGs

• Computer Science
AISTATS
• 2020
A completely general framework for learning sparse nonparametric directed acyclic graphs (DAGs) from data is developed that can be applied to general nonlinear models, general differentiable loss functions, and generic black-box optimization routines.

### 0-PENALIZED MAXIMUM LIKELIHOOD FOR SPARSE DIRECTED ACYCLIC GRAPHS BY SARA

It is shown that the 0-penalized maximum likelihood estimator of a DAG has about the same number of edges as the minimal-edge I-MAP (a DAG with minimal number of edge representing the distribution), and that it converges in Frobenius norm.

### DAG-GNN: DAG Structure Learning with Graph Neural Networks

• Yue YuJie Chen
• Computer Science
ICML
• 2019
A deep generative model is proposed and a variant of the structural constraint to learn the DAG is applied that learns more accurate graphs for nonlinearly generated samples; and on benchmark data sets with discrete variables, the learned graphs are reasonably close to the global optima.

### Joint skeleton estimation of multiple directed acyclic graphs for heterogeneous population

• Computer Science
Biometrics
• 2019
A two‐step approach to jointly estimate the DAG skeletons of multiple populations while the population origin of each sample may or may not be labeled, which allows a probabilistic soft label for each sample, which can be easily computed and often leads to more accurate skeleton estimation than hard labels.

### Causal Discovery with Reinforcement Learning

• Computer Science
ICLR
• 2020
This work proposes to use Reinforcement Learning (RL) to search for a Directed Acyclic Graph (DAG) according to a predefined score function and shows that the proposed approach not only has an improved search ability but also allows a flexible score function under the acyclicity constraint.