
Using Experimental Data to Evaluate Methods for Observational Causal Inference
Methods that infer causal dependence from observational data are central...
read it

From Dependence to Causation
Machine learning is the science of discovering statistical dependencies ...
read it

RealCause: Realistic Causal Inference Benchmarking
There are many different causal effect estimators in causal inference. H...
read it

A Causally Formulated Hazard Ratio Estimation through Backdoor Adjustment on Structural Causal Model
Identifying causal relationships for a treatment intervention is a funda...
read it

A Critical Look At The Identifiability of Causal Effects with Deep Latent Variable Models
Using deep latent variable models in causal inference has attracted cons...
read it

Causal inference under oversimplified longitudinal causal models
Most causal models of interest involve longitudinal exposures, confounde...
read it

Machine Learning for Public Administration Research, with Application to Organizational Reputation
Machine learning methods have gained a great deal of popularity in recen...
read it
The Case for Evaluating Causal Models Using Interventional Measures and Empirical Data
Causal inference is central to many areas of artificial intelligence, including complex reasoning, planning, knowledgebase construction, robotics, explanation, and fairness. An active community of researchers develops and enhances algorithms that learn causal models from data, and this work has produced a series of impressive technical advances. However, evaluation techniques for causal modeling algorithms have remained somewhat primitive, limiting what we can learn from experimental studies of algorithm performance, constraining the types of algorithms and model representations that researchers consider, and creating a gap between theory and practice. We argue for more frequent use of evaluation techniques that examine interventional measures rather than structural or observational measures, and that evaluate those measures on empirical data rather than synthetic data. We survey the current practice in evaluation and show that these are rarely used in practice. We show that such techniques are feasible and that data sets are available to conduct such evaluations. We also show that these techniques produce substantially different results than using structural measures and synthetic data.
READ FULL TEXT
Comments
There are no comments yet.