|
Activity Number:
|
132
- SLDS CSpeed 1
|
|
Type:
|
Contributed
|
|
Date/Time:
|
Monday, August 9, 2021 : 1:30 PM to 3:20 PM
|
|
Sponsor:
|
Section on Statistical Learning and Data Science
|
|
Abstract #318649
|
|
|
Title:
|
What Makes You Unique?
|
|
Author(s):
|
Ben Seiler* and Art Owen and Masayoshi Mase
|
|
Companies:
|
Stanford University and Stanford University and Hitachi, Ltd
|
|
Keywords:
|
uniqueness;
shapley values;
entropy
|
|
Abstract:
|
This paper proposes a uniqueness Shapley measure to compare the extent to which different variables are able to identify a subject. Revealing the value of a variable on subject t shrinks the set of possible subjects that t could be. The extent of the shrinkage depends on which other variables have also been revealed. We use Shapley value to combine the reductions in log cardinality due to revealing a variable given other subsets previously revealed. This uniqueness Shapley measure can be aggregated over subjects where it becomes a weighted sum of conditional entropies. Aggregation over subsets of subjects can address questions like how identifying is age for people of a given race. Such aggregates have a corresponding expression in terms of cross entropies. We use uniqueness Shapley to investigate the differential effects of revealing variables from the North Carolina voter registration rolls and in identifying anomalous solar flares. Our implementation relies on the ADtree data structure of Moore and Lee (1998) to store the cardinalities we need for fast computation.
|
Authors who are presenting talks have a * after their name.
Back to the full JSM 2021 program
|
