Differentially Private and Fair Deep Learning: A Lagrangian Dual Approach.

Cuong Tran Ferdinando Fioretto Pascal Van Hentenryck

Published in: CoRR (2020)

Keyphrases

deep learning
differentially private
lagrangian dual
cutting plane algorithm
lower and upper bounds
lagrangian relaxation
differential privacy

linear programming
duality gap
unsupervised learning
branch and bound
machine learning
linear program
optimal solution

dual variables
mental models
weakly supervised
lower bound
topic models
natural images
integer programming

upper bound
np hard
feature space
computer vision
learning algorithm
data sets