no code implementations • ICLR 2021 • Mark Niklas Mueller, Mislav Balunovic, Martin Vechev
In this work, we propose a new architecture which addresses this challenge and enables one to boost the certified robustness of any state-of-the-art deep network, while controlling the overall accuracy loss, without requiring retraining.
1 code implementation • 27 May 2020 • Wonryong Ryou, Jiayu Chen, Mislav Balunovic, Gagandeep Singh, Andrei Dan, Martin Vechev
We present a scalable and precise verifier for recurrent neural networks, called Prover based on two novel ideas: (i) a method to compute a set of polyhedral abstractions for the non-convex and nonlinear recurrent update functions by combining sampling, optimization, and Fermat's theorem, and (ii) a gradient descent based algorithm for abstraction refinement guided by the certification problem that combines multiple abstractions for each neuron.
1 code implementation • ICLR 2020 • Mislav Balunovic, Martin Vechev
We experimentally show that this training method, named convex layerwise adversarial training (COLT), is promising and achieves the best of both worlds -- it produces a state-of-the-art neural network with certified robustness of 60. 5% and accuracy of 78. 4% on the challenging CIFAR-10 dataset with a 2/255 L-infinity perturbation.
1 code implementation • NeurIPS 2019 • Mislav Balunovic, Maximilian Baader, Gagandeep Singh, Timon Gehr, Martin Vechev
The use of neural networks in safety-critical computer vision systems calls for their robustness certification against natural geometric transformations (e. g., rotation, scaling).
no code implementations • 25 Sep 2019 • Wonryong Ryou, Mislav Balunovic, Gagandeep Singh, Martin Vechev
We present the first end-to-end verifier of audio classifiers.
no code implementations • ICLR 2019 • Marc Fischer, Mislav Balunovic, Dana Drachsler-Cohen, Timon Gehr, Ce Zhang, Martin Vechev
We present DL2, a system for training and querying neural networks with logical constraints.
no code implementations • NeurIPS 2018 • Mislav Balunovic, Pavol Bielik, Martin Vechev
We present a new approach for learning to solve SMT formulas.