Explore behavioral analysis techniques for securing AI models against post-quantum threats. Learn how to identify anomalies and protect your AI infrastructure with quantum-resistant cryptography.
This is the official code repository of our BMVC 2023 paper "De-identification of facial videos while preserving remote physiological utility". Our method can de-identify videos from machines, while ...
Abstract: As a commonly used model for anomaly detection, the autoencoder model for anomaly detection does not train the objective for extracted features, which is a downside of autoencoder model. In ...
In this work, I explore methods for improving the cross-entropy loss alignment between SAE-reconstructed and original model logits. I find that: Rather than combining everything into one large and ...
Some results have been hidden because they may be inaccessible to you
Show inaccessible results