Adversarial robustness certification for Bayesian neural networks

Adversarial robustness certification for Bayesian neural networks

We study the problem of certifying the robustness of Bayesian neural networks (BNNs) to adversarial input perturbations. Specifically, we define two notions of robustness for BNNs in an adversarial setting: probabilistic robustness and decision robustness. The former deals with the probabilistic beh...

Detaylı Bibliyografya
Asıl Yazarlar:	Wicker, M, Patane, A, Laurenti, L, Kwiatkowska, M
Materyal Türü:	Conference item
Dil:	English
Baskı/Yayın Bilgisi:	Springer 2024

Benzer Materyaller

Certification of iterative predictions in Bayesian neural networks
Yazar:: Wicker, M, ve diğerleri
Baskı/Yayın Bilgisi: (2021)

Bayesian inference with certifiable adversarial robustness
Yazar:: Wicker, M, ve diğerleri
Baskı/Yayın Bilgisi: (2021)

Statistical guarantees for the robustness of Bayesian neural networks
Yazar:: Cardelli, L, ve diğerleri
Baskı/Yayın Bilgisi: (2019)

Adversarial robustness of Bayesian neural networks
Yazar:: Wicker, M
Baskı/Yayın Bilgisi: (2021)

Probabilistic safety for bayesian neural networks
Yazar:: Wicker, M, ve diğerleri
Baskı/Yayın Bilgisi: (2020)

Robustness of Bayesian neural networks to gradient-based attacks
Yazar:: Carbone, G, ve diğerleri
Baskı/Yayın Bilgisi: (2020)

Probabilistic reach-avoid for Bayesian neural networks
Yazar:: Wicker, M, ve diğerleri
Baskı/Yayın Bilgisi: (2024)

Adversarial robustness guarantees for Gaussian processes
Yazar:: Patane, A, ve diğerleri
Baskı/Yayın Bilgisi: (2022)

Adversarial robustness guarantees for classification with Gaussian Processes
Yazar:: Blaas, A, ve diğerleri
Baskı/Yayın Bilgisi: (2020)

Robustness guarantees for Bayesian inference with Gaussian processes
Yazar:: Cardelli, L, ve diğerleri
Baskı/Yayın Bilgisi: (2019)

Individual fairness guarantees for neural networks
Yazar:: Benussi, E, ve diğerleri
Baskı/Yayın Bilgisi: (2022)

Robustness of 3D deep learning in an adversarial setting
Yazar:: Wicker, M, ve diğerleri
Baskı/Yayın Bilgisi: (2020)

On the adversarial robustness of Gaussian processes
Yazar:: Patanè, A
Baskı/Yayın Bilgisi: (2020)

Enhancing adversarial robustness of deep neural networks
Yazar:: Zhang, Jeffrey,M. Eng.Massachusetts Institute of Technology.
Baskı/Yayın Bilgisi: (2019)

Tight certificates of adversarial robustness for randomly smoothed classifiers
Yazar:: Lee, Guang-He, ve diğerleri
Baskı/Yayın Bilgisi: (2021)

Robustness guarantees for Bayesian neural networks (invited extended abstract of a keynote speaker)
Yazar:: Kwiatkowska, M
Baskı/Yayın Bilgisi: (2022)

On the adversarial robustness of Bayesian machine learning models
Yazar:: Blaas, AC
Baskı/Yayın Bilgisi: (2021)

Adversarial Robustness Guarantees for Random Deep Neural Networks
Yazar:: De Palma, Giacomo, ve diğerleri
Baskı/Yayın Bilgisi: (2022)

Towards deep neural networks robust to adversarial examples
Yazar:: Matyasko, Alexander
Baskı/Yayın Bilgisi: (2020)

Fastened CROWN: Tightened Neural Network Robustness Certificates
Yazar:: Lyu, Zhaoyang, ve diğerleri
Baskı/Yayın Bilgisi: (2022)

When to trust AI: advances and challenges for certification of neural networks
Yazar:: Kwiatkowska, M, ve diğerleri
Baskı/Yayın Bilgisi: (2023)

Exploring the vulnerabilities and enhancing the adversarial robustness of deep neural networks
Yazar:: Bai, Tao
Baskı/Yayın Bilgisi: (2022)

Improved methodology for evaluating adversarial robustness in deep neural networks
Yazar:: Lee, Kyungmi,(Computer scientist)Massachusetts Institute of Technology.
Baskı/Yayın Bilgisi: (2020)

Robust full Bayesian methods for neural networks
Yazar:: Andrieu, C, ve diğerleri
Baskı/Yayın Bilgisi: (2000)

Efficient Neural Network Robustness Certification with General Activation Functions
Yazar:: Zhang, Huan, ve diğerleri
Baskı/Yayın Bilgisi: (2021)

Calibrating the classifier: siamese neural network architecture for end-to-end arousal recognition from ECG
Yazar:: Patane, A, ve diğerleri
Baskı/Yayın Bilgisi: (2019)

Studies on neural networks: Information propagation at initialisation and robustness to adversarial examples
Yazar:: Ughi, G
Baskı/Yayın Bilgisi: (2022)

FullCert: deterministic end-to-end certification for training and inference of neural networks
Yazar:: Lorenz, T, ve diğerleri
Baskı/Yayın Bilgisi: (2024)

Robust load forecasting towards adversarial attacks via Bayesian learning
Yazar:: Zhou, Y, ve diğerleri
Baskı/Yayın Bilgisi: (2022)

Robustness guarantees for deep neural networks on videos
Yazar:: Kwiatkowska, M, ve diğerleri
Baskı/Yayın Bilgisi: (2020)

Adversarial examples in neural networks
Yazar:: Lim, Ruihong
Baskı/Yayın Bilgisi: (2024)

Understanding the Energy vs. Adversarial Robustness Trade-Off in Deep Neural Networks
Yazar:: Lee, Kyungmi, ve diğerleri
Baskı/Yayın Bilgisi: (2022)

Adversarial Learned Soups: neural network averaging for joint clean and robust performance
Yazar:: Huang, Brian
Baskı/Yayın Bilgisi: (2023)

Feature-guided black-box safety testing of deep neural networks
Yazar:: Wicker, M, ve diğerleri
Baskı/Yayın Bilgisi: (2018)

Towards certification of uncertainty calibration under adversarial attacks
Yazar:: Emde, C, ve diğerleri
Baskı/Yayın Bilgisi: (2024)

Automated recognition of sleep arousal using multimodal and personalized deep ensembles of neural networks
Yazar:: Patane, A, ve diğerleri
Baskı/Yayın Bilgisi: (2019)

Adversarial attack defenses for neural networks
Yazar:: Puah, Yi Hao
Baskı/Yayın Bilgisi: (2024)

Demystifying adversarial attacks on neural networks
Yazar:: Yip, Lionell En Zhi
Baskı/Yayın Bilgisi: (2020)

Adversarial attack defences for neural network
Yazar:: Singh Kirath
Baskı/Yayın Bilgisi: (2022)

Robustness guarantees for credal Bayesian networks via constraint relaxation over probabilistic circuits
Yazar:: Wijk, H, ve diğerleri
Baskı/Yayın Bilgisi: (2022)