Search Results for author: Nikolaos Tsilivis

Found 3 papers, 1 papers with code

Can we achieve robustness from data alone?

1 code implementation24 Jul 2022 Nikolaos Tsilivis, Jingtong Su, Julia Kempe

Adversarial training and its variants have come to be the prevailing methods to achieve adversarially robust classification using neural networks.

Meta-Learning Robust classification

Extracting Finite Automata from RNNs Using State Merging

no code implementations28 Jan 2022 William Merrill, Nikolaos Tsilivis

One way to interpret the behavior of a blackbox recurrent neural network (RNN) is to extract from it a more interpretable discrete computational model, like a finite state machine, that captures its behavior.

The NTK Adversary: An Approach to Adversarial Attacks without any Model Access

no code implementations29 Sep 2021 Nikolaos Tsilivis, Julia Kempe

In particular, in the regime where the Neural Tangent Kernel theory holds, we derive a simple, but powerful strategy for attacking models, which in contrast to prior work, does not require any access to the model under attack, or any trained replica of it for that matter.

Learning Theory

Cannot find the paper you are looking for? You can Submit a new open access paper.