Causal Artificial Intelligence

Glossary

advanced analytics

Advanced analytics refers to the use of sophisticated techniques and tools to analyze and interpret complex data sets, uncovering valuable insights and patterns that can inform business decisions and drive strategic actions.

algorithm

This is a set of rules or instructions given to an AI, neural network, or other machine to help it learn on its own.

augmented intelligence

Augmented intelligence, also known as intelligence amplification, refers to the integration of artificial intelligence (AI) technologies and capabilities with human intelligence to enhance human decision-making, problem-solving, and cognitive abilities.

backdoor criterion

This is a method for identifying and controlling for confounding variables in a DAG, used in causal inference.

backdoor paths

In causal inference, a backdoor path refers to a noncausal pathway between a treatment variable and an outcome variable that can introduce bias or confounding when estimating the causal effect. It occurs when there is an alternative route, or a set of variables, that connects the treatment and outcome variables without involving any variables that lie on the causal pathway. Controlling for or conditioning on variables along a backdoor path can help eliminate or mitigate the confounding effect and estimate the true causal relationship between the treatment and outcome variables.

bias (in machine learning)

This is an error introduced in your model due to oversimplification of a machine ...