Jianwei Shuai's team and Jiahuai Han's team at Xiamen University have developed a deep autoencoder-based data-independent acquisition data analysis software for protein mass spectrometry, which ...

However, most existing autoencoder-based methods discard the reconstruction of auxiliary information, which poses a huge challenge for better representation learning and model scalability.

We propose an unsupervised method for detecting adversarial attacks in inner layers of autoencoder (AE) networks by maximizing a non-parametric measure of anomalous node activations.