HAI Weekly Seminar with Art Owen
Variable Importance, Cohort Shapley Value, and Redlining
Get the latest news, advances in research, policy work, and education program updates from HAI in your inbox weekly.
Sign Up For Latest News
Variable Importance, Cohort Shapley Value, and Redlining
AI+Science: Accelerating Discovery is an interdisciplinary conference bringing together researchers across physics, mathematics, chemistry, biology, neuroscience, and more to examine how AI is reshaping scientific discovery.

AI+Science: Accelerating Discovery is an interdisciplinary conference bringing together researchers across physics, mathematics, chemistry, biology, neuroscience, and more to examine how AI is reshaping scientific discovery.
While Large Language Models (LLMs) show promise in many domains, relying on them for direct policy generation in games often results in illegal moves and poor strategic play.

While Large Language Models (LLMs) show promise in many domains, relying on them for direct policy generation in games often results in illegal moves and poor strategic play.
The AI Index, currently in its ninth year, tracks, collates, distills, and visualizes data relating to artificial intelligence.

The AI Index, currently in its ninth year, tracks, collates, distills, and visualizes data relating to artificial intelligence.
In order to explain what a black box algorithm does, Owen says, we can start by studying which variables are important for its decisions. Variable importance is studied by making hypothetical changes to predictor variables. Changing parameters one at a time can produce input combinations that are outliers or very unlikely. They can be physically impossible, or even logically impossible. It is problematic to base an explanation on outputs corresponding to impossible inputs. Owen introduces the cohort Shapley (CS) measure to avoid this problem, based on Shapley value from cooperative game theory.
There are many tradeoffs in picking a variable importance measure, so CS is not the unique reasonable choice. One interesting property of CS is that it can detect 'redlining', meaning the impact of a protected variable on an algorithm's output when that algorithm was trained without the protected variable.