A low-dimensional voice latent space derived from deep learning captures speaker-identity representations in the temporal voice areas and supports reconstruction of voices preserving identity ...

Abstract: In industrial scenarios, the scarcity of labeled data for fault diagnosis of rotating machinery poses a significant challenge to the development of reliable data-driven models. This paper ...

Abstract: In recent years, the deployment of multiple Unmanned Aerial Vehicles (UAVs) for source localization has significantly increased in military, civil, and commercial domains. The transmission ...

Detect unknown (zero-day) IoT attacks using unsupervised anomaly detection with deep autoencoders. Train only on normal traffic - no attack labels needed!

We present Representation Autoencoders (RAE), a class of autoencoders that utilize pretrained, frozen representation encoders such as DINOv2 and SigLIP2 as encoders with trained ViT decoders. RAE can ...