Individual fairness concerns the ability of a machine learning model to not being affected in its predictions by one or more sensitive features, such as gender, race, age etc.
Recent methods developed techniques for the formal analysis and approximation of fairness in the case of deep Neural Networks (NNs). However such techniques are restricted to simple classification problems and cannot be applied to repeated procedures such as turned-base games, multi-step interviews etc.
This project will look into developing approximation methods for the analysis of individual fairness in Neural Network through time.
A familiarity with neural networks and with python is expected from the student.