DADI: Dynamic Discovery of Fair Information with Adversarial Reinforcement Learning

Bakker Michiel A.
Bakker Michiel A.
Tu Duy Patrick
Tu Duy Patrick
Valdés Humberto Riverón
Valdés Humberto Riverón
Varshney Kush R.
Varshney Kush R.
Pentland Alex
Pentland Alex
Cited by: 1|Bibtex|Views58
Other Links: arxiv.org

Abstract:

We introduce a framework for dynamic adversarial discovery of information (DADI), motivated by a scenario where information (a feature set) is used by third parties with unknown objectives. We train a reinforcement learning agent to sequentially acquire a subset of the information while balancing accuracy and fairness of predictors down...More

Code:

Data:

Full Text
Your rating :
0

 

Tags
Comments