Interpretable Clustering with Adaptive Heterogeneous Causal Structure Learning in Mixed Observational Data

2025-09-04

AI Summary

Existing methods struggle to identify causal heterogeneity without prior knowledge (e.g., temporal order or environmental labels) and inadequately model confounding effects and observational constraints, leading to poor interpretability and spurious associations. To address this, we propose HCL—a heterogeneous causal learning framework that jointly models structural heterogeneity and confounding via equivalent representation learning. HCL introduces bidirectional iterative optimization coupled with self-supervised regularization to enable synergistic causal clustering and heterogeneous causal graph discovery. We theoretically establish the identifiability of heterogeneous causal structures under relaxed assumptions—specifically, without requiring causal homogeneity or full confounder observability—and support mixed-type observational data. Experiments demonstrate that HCL achieves state-of-the-art performance on both causal clustering and causal structure recovery tasks. Moreover, applied to real single-cell perturbation data, HCL successfully recovers biologically meaningful causal mechanisms.

Technology Category

• Machine Learning: Causal Learning
• Reasoning under Uncertainty: Causality
• Constraint Satisfaction and Optimization: Constraint Learning and Acquisition

Application Category

• Graph Algorithms and Modeling for the Web: Representation, reconstruction, and subgraph or motif discovery in Web-related graphs
• Search and Retrieval-Augmented AI: Web learning to rank, online learning, and counterfactual learning for ranking
• Semantics and Knowledge: Methods, algorithms and applications for the development of semantic models, knowledge graphs and other forms of structured data models with machine-interpretable semantics

Abstract

Understanding causal heterogeneity is essential for scientific discovery in domains such as biology and medicine. However, existing methods lack causal awareness, with insufficient modeling of heterogeneity, confounding, and observational constraints, leading to poor interpretability and difficulty distinguishing true causal heterogeneity from spurious associations. We propose an unsupervised framework, HCL (Interpretable Causal Mechanism-Aware Clustering with Adaptive Heterogeneous Causal Structure Learning), that jointly infers latent clusters and their associated causal structures from mixed-type observational data without requiring temporal ordering, environment labels, interventions or other prior knowledge. HCL relaxes the homogeneity and sufficiency assumptions by introducing an equivalent representation that encodes both structural heterogeneity and confounding. It further develops a bi-directional iterative strategy to alternately refine causal clustering and structure learning, along with a self-supervised regularization that balance cross-cluster universality and specificity. Together, these components enable convergence toward interpretable, heterogeneous causal patterns. Theoretically, we show identifiability of heterogeneous causal structures under mild conditions. Empirically, HCL achieves superior performance in both clustering and structure learning tasks, and recovers biologically meaningful mechanisms in real-world single-cell perturbation data, demonstrating its utility for discovering interpretable, mechanism-level causal heterogeneity.

Problem

• Identifies causal heterogeneity in mixed observational data
• Overcomes limitations in modeling heterogeneity and confounding
• Recovers interpretable causal patterns without prior knowledge

Innovation

• Adaptive heterogeneous causal structure learning
• Bi-directional iterative clustering strategy
• Self-supervised regularization balancing universality specificity