DAG Learning from Zero-Inflated Count Data Using Continuous Optimization
This addresses network inference problems in domains like genomics where data often have many zeros, offering a scalable solution for researchers in computational biology.
The paper tackles learning directed acyclic graph structures from zero-inflated count data by proposing ZICO, a method that uses continuous optimization with a differentiable acyclicity constraint, achieving superior performance and faster runtimes on simulated data and comparable or better results on gene regulatory networks.
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.