4 papers
On Adversarial Vulnerability of Vision-Language Models through the Lens of Intermediate Spectral Subspaces
Chethan Krishnamurthy Ramanaik, Tobias Callies, Michael Hecht +1
Adversarial vulnerability in deep neural networks (DNNs) has been studied from the perspectives of decision-boundary geometry, feature robustness, input-output Jacobians, and the i…
Revealing Hidden Vulnerabilities in Autoencoders through Gradient Signal Restoration
Chethan Krishnamurthy Ramanaik, Arjun Roy, Tobias Callies +1
Adversarial robustness of deep autoencoders (AEs) has received less attention than that of discriminative models, although their compressed latent representations induce ill-condit…
Adversarial Robustness of VAEs across Intersectional Subgroups
Chethan Krishnamurthy Ramanaik, Arjun Roy, Eirini Ntoutsi
Despite advancements in Autoencoders (AEs) for tasks like dimensionality reduction, representation learning and data generation, they remain vulnerable to adversarial attacks. Vari…
Ensuring Topological Data-Structure Preservation under Autoencoder Compression due to Latent Space Regularization in Gauss--Legendre nodes
Chethan Krishnamurthy Ramanaik, Juan-Esteban Suarez Cardona, Anna Willmann +3
We formulate a data independent latent space regularisation constraint for general unsupervised autoencoders. The regularisation rests on sampling the autoencoder Jacobian in Legen…