Abstract: Missing data can pose severe consequences in critical contexts, such as clinical research based on routinely collected healthcare data. This issue is usually handled with imputation ...

Abstract: Autoencoder (AE) is a deep neural network that has been widely utilized in process industry owing to its superior abilities of feature extraction and data reconstruction. Recently, assuming ...

Graph neural network (GNN) autoencoders such as STAGATE and GraphST are the state of the art for spatial domain identification, but their results are surprisingly sensitive to the random seed: ...

Detecting Adversarial Evasion Attacks Against Autoencoder-Based Network Intrusion Detection Systems Niklas Bunzel, Ashim Siwakoti Generative AI and Federated Learning for Intrusion Detection Systems: ...