The success of machine learning is fueled by the increasing availability of computing power
and large training datasets. The training data is used to learn new models or update existing …

Semi-supervised learning (SSL) has tremendous value in practice due to the utilization of
both labeled and unlabelled data. An essential class of SSL methods, referred to as graph …

The field of graph neural networks (GNNs) has seen rapid and incredible strides over the
recent years. Graph neural networks, also known as deep learning on graphs, graph …

Graph Neural Networks (GNNs) are powerful tools in representation learning for graphs.
However, recent studies show that GNNs are vulnerable to carefully-crafted perturbations …

Deep learning methods for graphs achieve remarkable performance on many tasks.
However, despite the proliferation of such methods and their success, recent findings …

Multimodal contrastive learning methods like CLIP train on noisy and uncurated training
datasets. This is cheaper than labeling datasets manually, and even improves out-of …

Adversarial Attacks and Defenses on Graphs Page 1 Adversarial Attacks and Defenses on
Graphs: A Review, A Tool and Empirical Studies Wei Jin†, Yaxin Li†, Han Xu†, Yiqi Wang† …

Backdoor attack intends to inject hidden backdoor into the deep neural networks (DNNs),
such that the prediction of the infected model will be maliciously changed if the hidden …

One intriguing property of deep neural networks (DNNs) is their inherent vulnerability to
backdoor attacks—a trojan model responds to trigger-embedded inputs in a highly …

Semi-supervised machine learning models learn from a (small) set of labeled training
examples, and a (large) set of unlabeled training examples. State-of-the-art models can …