DAG Learning from Zero-Inflated Count Data Using Continuous Optimization

Noriaki Sato, Marco Scutari, Shuichi Kawano, Rui Yamaguchi, Seiya Imoto
Proceedings of the Fifth Conference on Causal Learning and Reasoning, PMLR 323:277-296, 2026.

Abstract

We address network structure learning from zero-inflated count data by casting each node as a zero-inflated generalized linear model and optimizing a smooth, score-based objective under a directed acyclic graph constraint. Our Zero-Inflated Continuous Optimization (ZICO) approach uses node-wise likelihoods with canonical links and enforces acyclicity through a differentiable surrogate constraint combined with sparsity regularization. ZICO achieves superior performance with faster runtimes on simulated data. It also performs comparably to or better than common algorithms for reverse engineering gene regulatory networks. ZICO is fully vectorized and mini-batched, enabling learning on larger variable sets with practical runtimes in a wide range of domains.

Cite this Paper

BibTeX
@InProceedings{pmlr-v323-sato26a, title = {DAG Learning from Zero-Inflated Count Data Using Continuous Optimization}, author = {Sato, Noriaki and Scutari, Marco and Kawano, Shuichi and Yamaguchi, Rui and Imoto, Seiya}, booktitle = {Proceedings of the Fifth Conference on Causal Learning and Reasoning}, pages = {277--296}, year = {2026}, editor = {Mazaheri, Bijan and Hanson, Niels Richard}, volume = {323}, series = {Proceedings of Machine Learning Research}, month = {06--08 Apr}, publisher = {PMLR}, pdf = {https://raw.githubusercontent.com/mlresearch/v323/main/assets/sato26a/sato26a.pdf}, url = {https://proceedings.mlr.press/v323/sato26a.html}, abstract = {We address network structure learning from zero-inflated count data by casting each node as a zero-inflated generalized linear model and optimizing a smooth, score-based objective under a directed acyclic graph constraint. Our Zero-Inflated Continuous Optimization (ZICO) approach uses node-wise likelihoods with canonical links and enforces acyclicity through a differentiable surrogate constraint combined with sparsity regularization. ZICO achieves superior performance with faster runtimes on simulated data. It also performs comparably to or better than common algorithms for reverse engineering gene regulatory networks. ZICO is fully vectorized and mini-batched, enabling learning on larger variable sets with practical runtimes in a wide range of domains.} }
Endnote
%0 Conference Paper %T DAG Learning from Zero-Inflated Count Data Using Continuous Optimization %A Noriaki Sato %A Marco Scutari %A Shuichi Kawano %A Rui Yamaguchi %A Seiya Imoto %B Proceedings of the Fifth Conference on Causal Learning and Reasoning %C Proceedings of Machine Learning Research %D 2026 %E Bijan Mazaheri %E Niels Richard Hanson %F pmlr-v323-sato26a %I PMLR %P 277--296 %U https://proceedings.mlr.press/v323/sato26a.html %V 323 %X We address network structure learning from zero-inflated count data by casting each node as a zero-inflated generalized linear model and optimizing a smooth, score-based objective under a directed acyclic graph constraint. Our Zero-Inflated Continuous Optimization (ZICO) approach uses node-wise likelihoods with canonical links and enforces acyclicity through a differentiable surrogate constraint combined with sparsity regularization. ZICO achieves superior performance with faster runtimes on simulated data. It also performs comparably to or better than common algorithms for reverse engineering gene regulatory networks. ZICO is fully vectorized and mini-batched, enabling learning on larger variable sets with practical runtimes in a wide range of domains.
APA
Sato, N., Scutari, M., Kawano, S., Yamaguchi, R. & Imoto, S.. (2026). DAG Learning from Zero-Inflated Count Data Using Continuous Optimization. Proceedings of the Fifth Conference on Causal Learning and Reasoning, in Proceedings of Machine Learning Research 323:277-296 Available from https://proceedings.mlr.press/v323/sato26a.html.

Related Material