Convolutional layers have long served as the primary workhorse for image...
Differential privacy (DP) is the prevailing technique for protecting use...
LASSO regularized logistic regression is particularly useful for its bui...
Linear L_1-regularized models have remained one of the simplest and most...
Subsampling algorithms are a natural approach to reduce data size before...
We present a framework to statistically audit the privacy guarantee conf...
Many metric learning tasks, such as triplet learning, nearest neighbor
r...
We explore the utility of information contained within a dropout based
B...
Ordinal regression is a classification task where classes have an order ...
Deep neural networks (DNN) have been used successfully in many scientifi...
Learning disentangled representations is regarded as a fundamental task ...
Ambitious global goals have been established to provide universal access...