A Provable Defense for Deep Residual Networks
Not on the map. Why?
Abstract (arXiv)
We present a training system, which can provably defend significantly larger neural networks than previously possible, including ResNet-34 and DenseNet-100. Our approach is based on differentiable abstract interpretation and introduces two novel concepts: (i) abstract layers for fine-tuning the precision and scalability of the abstraction, (ii) a flexible domain specific language (DSL) for describing training objectives that combine abstract and concrete losses with arbitrary specifications. Our training method is implemented in the DiffAI system.
Key facts. Cited by 5 papers in the map (data through 14 Oct 2026). First posted on arXiv 29 Mar 2019.
Builds on
No reference list yet.
Built on by
Citing papers per year: 2021: 1; 2022: 1; 2023: 1; 2024: 2.
5 papers.
- 2021Towards Certifying L-infinity Robustness using Neural Networks with L-inf-dist Neuronscited by 40
- 2024Neural Network Verification with PyRATcited by 9
- 2022Keep your Distance: Determining Sampling and Distance Thresholds in Machine Learning Monitoringcited by 4
- 2023Scope Compliance Uncertainty Estimatecited by 3
- 2024Certified Training with Branch-and-Bound for Lyapunov-stable Neural Controlcited by 3
- 2024Certified Training with Branch-and-Bound for Lyapunov-stable Neural Control27 Nov 2024
- 2024Neural Network Verification with PyRAT31 Oct 2024
- 2023Scope Compliance Uncertainty Estimate17 Dec 2023
- 2022Keep your Distance: Determining Sampling and Distance Thresholds in Machine Learning Monitoring11 Jul 2022
- 2021Towards Certifying L-infinity Robustness using Neural Networks with L-inf-dist Neurons10 Feb 2021