Andrea L. Bertozzi
3 papers in the PaperMetrix corpus
Papers by this author
-
Adversarial Defense via Data Dependent Activation Function and Total Variation Minimization
2018 · arXiv (Cornell University)
We improve the robustness of Deep Neural Net (DNN) to adversarial attacks by using an interpolating function as the output activation. This data-dependent activation remarkably improves both the generalization and robustness of DNN. In the …
-
Efficient and Reliable Overlay Networks for Decentralized Federated Learning
2021 · arXiv (Cornell University)
We propose near-optimal overlay networks based on $d$-regular expander graphs to accelerate decentralized federated learning (DFL) and improve its generalization. In DFL a massive number of clients are connected by an overlay network, and they …
-
Efficient Graph-Based Active Learning with Probit Likelihood via\n Gaussian Approximations
2020 · arXiv (Cornell University)
We present a novel adaptation of active learning to graph-based\nsemi-supervised learning (SSL) under non-Gaussian Bayesian models. We present\nan approximation of non-Gaussian distributions to adapt previously\nGaussian-based acquisition functions to these more general cases. We develop an\nefficient …