structural-causal-models

DrugTargetWorld: a synthetic biobank for training and benchmarking AI scientists on end-to-end drug target discovery. Paper: arXiv:2610.09558

GenPark AI Agent Skill - Structural Causal Model (SCM) DAG engine evaluating observational distributions and simulating Pearl's do-calculus interventions.

GenPark AI Agent Skill - Structural Causal Model (SCM) DAG engine evaluating observational distributions and simulating Pearl's do-calculus interventions.

GenPark AI Agent Skill - Propensity score matching and Inverse Probability Weighting (IPW) estimator calculating Average Treatment Effect (ATE) under conditional ignorability.

GenPark AI Agent Skill - Pearl's 3-step counterfactual inference engine executing Abduction, Action (do-surgery), and Prediction to answer what-if causal inquiries.

GenPark AI Agent Skill - Graph-theoretic backdoor criterion validator and minimal confounder adjustment set identifier ensuring unconfounded causal effect estimation.

GenPark AI Agent Skill - Instrumental Variable (IV) Two-Stage Least Squares (2SLS) causal estimator isolating unobserved confounding and testing instrument strength.

GenPark AI Agent Skill - Instrumental Variable (IV) Two-Stage Least Squares (2SLS) causal estimator isolating unobserved confounding and testing instrument strength.

GenPark AI Agent Skill - Pearl's 3-step counterfactual inference engine executing Abduction, Action (do-surgery), and Prediction to answer what-if causal inquiries.

GenPark AI Agent Skill - Propensity score matching and Inverse Probability Weighting (IPW) estimator calculating Average Treatment Effect (ATE) under conditional ignorability.

GenPark AI Agent Skill - Graph-theoretic backdoor criterion validator and minimal confounder adjustment set identifier ensuring unconfounded causal effect estimation.

Air-gapped, edge-native industrial automation platform featuring Structural Causal Model (SCM) fault diagnosis, lock-free Rust execution, zero-downtime ONNX model hot-swapping, Ed25519 cryptographic security, and automated SAP/Maximo CMMS integration.

Efficient and diverse synthetic causal priors in JAX for end-to-end pretraining

Models causals estructurals i resultats potencials (Treball de Final de Grau)

Paper and Lean 4 mechanization for Graph Surgery and the Do-Operator

Dokumentasi lengkap Tugas 3 Causal Inference - 6 latihan menggunakan R & RStudio mencakup DAG, Simpson's Paradox, Backdoor Adjustment, Counterfactual, CausalImpact, dan Mediation Analysis

A PyTorch-native Bayesian network library where each mechanism is learnt by a neural network. Every node carries a learnable, batched, GPU-resident conditional distribution; every query is a batched tensor operation; inference and parameter learning both run end-to-end on gpu. https://nbn.readthedocs.io

Demystifying Judea Pearl's do-calculus and the Backdoor Criterion by hand. Resolving a curated Simpson's Paradox using structural causal modeling on superhero battle mechanics.

Intent Compilation, Multi-Intent Drift Detection, and Adaptive Delegation Envelopes for Autonomous Intent-Based Networking

Stress-testing causal discovery under hidden confounding using causalXtreme and extremeSCM, with reproducible R experiments on spurious causal edges and false-positive control.