awesome-causality-algorithms is a curated index of algorithms and tools for causal inference and causal machine learning.
The repository addresses the challenge of navigating the growing landscape of causality-focused algorithms by organizing methods into two primary categories: machine learning approaches that solve causal inference problems, and causal machine learning approaches that solve traditional machine learning problems using causal techniques. The index emphasizes reproducibility by including only methods with publicly available open-source code, except for survey and review papers. It provides structured access to toolboxes, treatment effect estimation methods, and causal machine learning algorithms, each with links to implementations and documentation.
Developers should use this index when building systems that require causal reasoning, treatment effect estimation, or causal discovery. It suits projects in areas like uplift modeling, heterogeneous treatment effect estimation, policy learning, and counterfactual inference. The repository organizes comprehensive toolboxes such as causal-learn, YLearn, and Trustworthy AI alongside specialized libraries for specific problems like DoWhy for causal inference with explicit assumption testing, EconML for individualized causal response estimation, and CausalML for uplift modeling. This structure helps practitioners identify which tool matches their specific causal inference need rather than treating all methods as interchangeable.
The project maintains an actively curated collection with a clear quality gate requiring open-source implementations. The index references a foundational survey paper and organizes entries with consistent metadata including code availability, documentation links, and brief descriptions of each method's focus. The emphasis on reproducibility through code availability and the structured categorization suggest ongoing maintenance focused on practical utility for practitioners seeking to apply causal methods.
