Membership inference attacks are designed to determine, using black box
...
We revisit the problem of differentially private squared error linear
re...
We provide a differentially private algorithm for producing synthetic da...
Large-scale e-commercial platforms in the real-world usually contain var...
The method of Common Spatial Patterns (CSP) is widely used for feature
e...
The inductive biases of trained neural networks are difficult to underst...
The large amount of online data and vast array of computing resources en...
Style transfer methods produce a transferred image which is a rendering ...
Word embeddings learnt from large corpora have been adopted in various
a...
idely used recurrent units, including Long-short Term Memory (LSTM) and ...
Widely used recurrent units, including Long-short Term Memory (LSTM) and...
Multi-view learning can provide self-supervision when different views ar...
The encoder-decoder models for unsupervised sentence representation lear...
Multi-view learning can provide self-supervision when different views ar...
Style transfer methods produce a transferred image which is a rendering ...
A popular series of style transfer methods apply a style to a content im...
Context information plays an important role in human language understand...
The skip-thought model has been proven to be effective at learning sente...
We study the skip-thought model with neighborhood information as weak
su...
Top-down information plays a central role in human perception, but plays...