Search Results for author:
Found 3 papers, 0 papers with code
Unveiling Privacy, Memorization, and Input Curvature Links
Second, we present a novel insight showing that input loss curvature is upper-bounded by the differential privacy parameter.
Synthetic Dataset Generation for Privacy-Preserving Machine Learning
We evaluate our method on image classification dataset (CIFAR10) and show that our synthetic data can be used for training networks from scratch, producing reasonable classification performance.
Evaluating the Stability of Recurrent Neural Models during Training with Eigenvalue Spectra Analysis
We analyze the stability of recurrent networks, specifically, reservoir computing models during training by evaluating the eigenvalue spectra of the reservoir dynamics.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
