Bayesian inference with certifiable adversarial robustness

Bayesian inference with certifiable adversarial robustness

We consider adversarial training of deep neural networks through the lens of Bayesian learning and present a principled framework for adversarial training of Bayesian Neural Networks (BNNs) with certifiable guarantees. We rely on techniques from constraint relaxation of non-convex optimisation probl...

Podrobná bibliografie
Hlavní autoři:	Wicker, M, Laurenti, L, Patane, A, Chen, Z, Zhang, Z, Kwiatkowska, M
Médium:	Conference item
Jazyk:	English
Vydáno:	Journal of Machine Learning Research 2021

Podobné jednotky

Adversarial robustness certification for Bayesian neural networks
Autor: Wicker, M, a další
Vydáno: (2024)

Robustness guarantees for Bayesian inference with Gaussian processes
Autor: Cardelli, L, a další
Vydáno: (2019)

Statistical guarantees for the robustness of Bayesian neural networks
Autor: Cardelli, L, a další
Vydáno: (2019)

Adversarial robustness guarantees for Gaussian processes
Autor: Patane, A, a další
Vydáno: (2022)

Probabilistic safety for bayesian neural networks
Autor: Wicker, M, a další
Vydáno: (2020)

Adversarial robustness of Bayesian neural networks
Autor: Wicker, M
Vydáno: (2021)

Adversarial robustness guarantees for classification with Gaussian Processes
Autor: Blaas, A, a další
Vydáno: (2020)

Robustness of Bayesian neural networks to gradient-based attacks
Autor: Carbone, G, a další
Vydáno: (2020)

Robustness of 3D deep learning in an adversarial setting
Autor: Wicker, M, a další
Vydáno: (2020)

Certification of iterative predictions in Bayesian neural networks
Autor: Wicker, M, a další
Vydáno: (2021)

Probabilistic reach-avoid for Bayesian neural networks
Autor: Wicker, M, a další
Vydáno: (2024)

Certified Robustness to Text Adversarial Attacks by Randomized [MASK]
Autor: Jiehang Zeng, a další
Vydáno: (2023-06-01)

On the adversarial robustness of Gaussian processes
Autor: Patanè, A
Vydáno: (2020)

Certifiable Robustness to Adversarial State Uncertainty in Deep Reinforcement Learning
Autor: Everett, Michael F, a další
Vydáno: (2021)

Certifiable Robustness to Adversarial State Uncertainty in Deep Reinforcement Learning
Autor: Everett, Michael, a další
Vydáno: (2021)

Individual fairness guarantees for neural networks
Autor: Benussi, E, a další
Vydáno: (2022)

Robust load forecasting towards adversarial attacks via Bayesian learning
Autor: Zhou, Y, a další
Vydáno: (2022)

On the adversarial robustness of Bayesian machine learning models
Autor: Blaas, AC
Vydáno: (2021)

Reachability computation for switching diffusions: finite abstractions with certifiable and tuneable precision
Autor: Laurenti, L, a další
Vydáno: (2017)

Reachability computation for switching diffusions: Finite abstractions with certifiable and tuneable precision
Autor: Laurenti, L, a další
Vydáno: (2017)

Time Series Forecasting with Missing Data Using Generative Adversarial Networks and Bayesian Inference
Autor: Xiaoou Li
Vydáno: (2024-04-01)

Adversarially Robust Fault Zone Prediction in Smart Grids With Bayesian Neural Networks
Autor: Emad Efatinasab, a další
Vydáno: (2024-01-01)

Trustworthiness and certified robustness for deep learning
Autor: Xia, Song
Vydáno: (2022)

Certifiable Outlier-Robust Geometric Perception
Autor: Yang, Heng
Vydáno: (2023)

Robustness guarantees for Bayesian neural networks (invited extended abstract of a keynote speaker)
Autor: Kwiatkowska, M
Vydáno: (2022)

Certifiers make neural networks vulnerable to availability attacks
Autor: Lorenz, T, a další
Vydáno: (2023)

Robust simulation-based inference in cosmology with Bayesian neural networks
Autor: Pablo Lemos, a další
Vydáno: (2023-01-01)

Adversarial Inference Control in Cyber-Physical Systems: A Bayesian Approach With Application to Smart Meters
Autor: Ramana R. Avula, a další
Vydáno: (2024-01-01)

Inferring adversarial behaviour in cyber‐physical power systems using a Bayesian attack graph approach
Autor: Abhijeet Sahu, a další
Vydáno: (2023-06-01)

AutoBayes: Automated Bayesian Graph Exploration for Nuisance- Robust Inference
Autor: Andac Demir, a další
Vydáno: (2021-01-01)

Adversarial attacks and adversarial robustness in computational pathology
Autor: Narmin Ghaffari Laleh, a další
Vydáno: (2022-09-01)

Towards Fast computation of certified robustness for ReLU networks
Autor: Weng, Tsui-Wei, a další
Vydáno: (2021)

ME-Net: Towards effective adversarial robustness with matrix estimation
Autor: Yang, Y, a další
Vydáno: (2021)

Bayesian inference in FMRI.
Autor: Woolrich, M
Vydáno: (2012)

Bayesian inference in FMRI
Autor: Woolrich, M
Vydáno: (2012)

On the adversarial robustness of aerial detection
Autor: Yuwei Chen, a další
Vydáno: (2024-11-01)

Multivariate Financial Time-Series Prediction With Certified Robustness
Autor: Hui Li, a další
Vydáno: (2020-01-01)

Robustness guarantees for credal Bayesian networks via constraint relaxation over probabilistic circuits
Autor: Wijk, H, a další
Vydáno: (2022)

Manifold-driven decomposition for adversarial robustness
Autor: Wenjia Zhang, a další
Vydáno: (2024-01-01)

The R Package MitISEM: Efficient and Robust Simulation Procedures for Bayesian Inference
Autor: Nalan Baştürk, a další
Vydáno: (2017-07-01)