Efficient Certification of Spatial Robustness

02/02/2021

Efficient Certification of Spatial Robustness

Anian Ruoss, Maximilian Baader, Mislav Balunović, Martin Vechev

Keywords:

Abstract Paper Similar Papers

Abstract: Recent work has exposed the vulnerability of computer vision models to vector field attacks. Due to the widespread usage of such models in safety-critical applications, it is crucial to quantify their robustness against such spatial transformations. However, existing work only provides empirical robustness quantification against vector field deformations via adversarial attacks, which lack provable guarantees. In this work, we propose novel convex relaxations, enabling us, for the first time, to provide a certificate of robustness against vector field transformations. Our relaxations are model-agnostic and can be leveraged by a wide range of neural network verifiers. Experiments on various network architectures and different datasets demonstrate the effectiveness and scalability of our method.

The video of this talk cannot be embedded. You can watch it here:

https://slideslive.com/38949145

(Link will open in new window)

0

1

1

0

Share

This is an embedded video. Talk and the respective paper are published at AAAI 2021 virtual conference. If you are one of the authors of the paper and want to manage your upload, see the question "My papertalk has been externally embedded..." in the FAQ section.

Comments

Post Comment

no comments yet

Similar Papers

03/05/2021

Enforcing robust control guarantees within neural network policies

Priya Donti, Melrose Roderick, Mahyar Fazlyab, Zico Kolter

Keywords Paper

reinforcement learning, differentiable optimization, robust control

0

0

0

1

5:09

14/06/2020

Boosting the Transferability of Adversarial Samples via Attention

Weibin Wu, Yuxin Su, Xixian Chen and
Shenglin Zhao, Irwin King, Michael R. Lyu, Yu-Wing Tai

Keywords Paper

adversarial learning, adversarial attack methods

0

0

0

0

1:01

13/04/2021

Online robust control of nonlinear systems with large uncertainty

Dimitar Ho, Hoang Le, John Doyle, Yisong Yue

Keywords Paper

0

0

0

0

3:02

14/06/2020

What Machines See Is Not What They Get: Fooling Scene Text Recognition Models With Adversarial Text Images

Xing Xu, Jiefu Chen, Jinhui Xiao and
Lianli Gao, Fumin Shen, Heng Tao Shen

Keywords Paper

adversarial attack, scene text recognition, white-box attack, targeted attack, untargeted attack

0

0

0

0

5:00

18/11/2020

Towards understanding and improving the transferability of adversarial examples in deep neural networks

Lei Wu, Zhanxing Zhu

Keywords Paper

0

0

0

0

10:28

14/06/2020

QEBA: Query-Efficient Boundary-Based Blackbox Attack

Huichen Li, Xiaojun Xu, Xiaolu Zhang and
Shuang Yang, Bo Li

Keywords Paper

adversarial machine learning, black-box attack, boundary-based attack, attacking public api

0

0

0

0

1:01

06/12/2020

Smoothed Geometry for Robust Attribution

Zifan Wang, Haofan Wang, Shakul Ramkumar and
Peter Mardziel, Matt Fredrikson, Anupam Datta

Keywords Paper

0

0

0

0

3:27

03/05/2021

Incorporating Symmetry into Deep Dynamics Models for Improved Generalization

Rui Wang, Robin Walters, Rose Yu

Keywords Paper

AI for earth science, physics-guided deep learning, equivariant neural network, deep sequence model

0

0

0

0

4:46

26/04/2020

Bridging Mode Connectivity in Loss Landscapes and Adversarial Robustness

Pu Zhao, Pin-Yu Chen, Payel Das and
Karthikeyan Natesan Ramamurthy, Xue Lin

Keywords Paper

mode connectivity, adversarial robustness, backdoor attack, error-injection attack, evasion attacks, loss landscapes

0

0

0

0

4:30

06/12/2021

Adversarial Robustness with Non-uniform Perturbations

Ecenaz Erdemir, Jeffrey Bickford, Luca Melis, Sergul Aydore

Keywords Paper

deep learning, optimization, machine learning, robustness, adversarial robustness and security

0

0

0

0

15:05

18/07/2021

Gaussian Process-Based Real-Time Learning for Safety Critical Applications

Armin Lederer, Alejandro Ordóñez Conejo, Korbinian Maier and
Wenxin Xiao, Jonas Umlauft, Sandra Hirche

Keywords Paper

Probabilistic Methods, Gaussian Processes and Bayesian non-parametrics

0

0

0

0

4:59

03/05/2021

Towards Robust Neural Networks via Close-loop Control

Zhuotong Chen, Qianxiao Li, Zheng Zhang

Keywords Paper

dynamical system, neural network robustness, optimal control

0

0

0

0

4:47

06/12/2021

Adversarially Robust 3D Point Cloud Recognition Using Self-Supervisions

Jiachen Sun, Yulong Cao, Christopher B Choy and
Zhiding Yu, Anima Anandkumar, Zhuoqing Morley Mao, Chaowei Xiao

Keywords Paper

deep learning, robustness, adversarial robustness and security, self-supervised learning, transformers

0

0

0

0

13:15

12/07/2020

Proper Network Interpretability Helps Adversarial Robustness in Classification

Akhilan Boopathy, Sijia Liu, Gaoyuan Zhang and
Cynthia Liu, Pin-Yu Chen, Shiyu Chang, Luca Daniel

Keywords Paper

Accountability, Transparency and Interpretability

0

0

0

0

15:01

06/12/2020

Robustness of Bayesian Neural Networks to Gradient-Based Attacks

Ginevra Carbone, Matthew Wicker, Luca Laurenti and
Andrea Patane', Luca Bortolussi, Guido Sanguinetti

Keywords Paper

0

0

0

0

3:08

14/06/2020

When NAS Meets Robustness: In Search of Robust Architectures Against Adversarial Attacks

Minghao Guo, Yuzhe Yang, Rui Xu and
Ziwei Liu, Dahua Lin

Keywords Paper

adversarial robustness, neural architecture search, adversarial examples, deep learning architectures, adversarial attacks

0

0

0

0

1:01

30/11/2020

Bridging Adversarial and Statistical Domain Transfer via Spectral Adaptation Networks

Christoph Raab, Philipp Väth, Peter Meier, Frank-Michael Schleif

Keywords Paper

0

0

0

0

10:07

06/12/2021

Stabilizing Deep Q-Learning with ConvNets and Vision Transformers under Data Augmentation

Nicklas Hansen, Hao Su, Xiaolong Wang

Keywords Paper

reinforcement learning and planning, transformers

0

0

0

0

8:43

12/07/2020

Adversarial Neural Pruning with Latent Vulnerability Suppression

Divyam Madaan, Jinwoo Shin, Sung Ju Hwang

Keywords Paper

Adversarial Examples

0

0

0

0

14:43

12/08/2020

SmartVerif: Push the Limit of Automation Capability of Verifying Security Protocols by Dynamic Strategies

Yan Xiong, Cheng Su, Wenchao Huang and
Fuyou Miao, Wansen Wang, Hengyi Ouyang

Keywords Paper

0

0

0

0

11:18

25/06/2020

SaFace: Towards Scenario-aware Face Recognition via Edge Computing System

Zhe Zhou, Bingzhe Wu, Zheng Liang and
Guangyu Sun, Chenren Xu, Guojie Luo

Keywords Paper

0

0

0

0

11:28

14/06/2020

Benchmarking Adversarial Robustness on Image Classification

Yinpeng Dong, Qi-An Fu, Xiao Yang and
Tianyu Pang, Hang Su, Zihao Xiao, Jun Zhu

Keywords Paper

adversarial robustness, benchmark, evaluation, security, attack, defense, image classification

0

0

0

0

4:59

03/05/2021

Uncertainty in Gradient Boosting via Ensembles

Andrey Malinin, Liudmila Prokhorenkova, Aleksei Ustimenko

Keywords Paper

uncertainty, knowledge uncertainty, decision trees, gradient boosting, ensembles

0

0

0

0

5:30

06/12/2020

HYDRA: Pruning Adversarially Robust Neural Networks

Vikash Sehwag, Shiqi Wang, Prateek Mittal, Suman Jana

Keywords Paper

0

0

0

0

3:14

12/07/2020

On Breaking Deep Generative Model-based Defenses and Beyond

Yanzhi Chen, Renjie Xie, Zhanxing Zhu

Keywords Paper

Adversarial Examples

0

0

0

0

10:33

22/11/2021

On Adversarial Robustness of 3D Point Cloud Classification under Adaptive Attacks

Jiachen Sun, Karl Koenig, Yulong Cao and
Qi Alfred Chen, Zhuoqing Morley Mao

Keywords Paper

point cloud classification, adversarial training, adaptive attack

0

0

0

0

3:05

02/02/2021

A Theory of Independent Mechanisms for Extrapolation in Generative Models

Michel Besserve, Remy Sun, Dominik Janzing, Bernhard Schölkopf

Keywords Paper

0

0

0

0

18:37

18/07/2021

Continuous-time Model-based Reinforcement Learning

Cagatay Yildiz, Markus Heinonen, Harri Lähdesmäki

Keywords Paper

Reinforcement Learning and Planning

0

0

0

0

5:00

06/12/2020

Maximum-Entropy Adversarial Data Augmentation for Improved Generalization and Robustness

Long Zhao, Ting Liu, Xi Peng, Dimitris Metaxas

Keywords Paper

0

0

0

0

3:22

14/09/2020

Robust Training of Graph Convolutional Networks via Latent Perturbation

Hongwei Jin, Xinhua Zhang

Keywords Paper

graph neural network, adversarial training, representation learning

0

0

0

0

17:38

06/12/2021

Integrated Latent Heterogeneity and Invariance Learning in Kernel Space

Jiashuo Liu, Zheyuan Hu, Peng Cui and
Bo Li, Zheyan Shen

Keywords Paper

deep learning, reinforcement learning and planning, machine learning

0

0

0

0

11:11

03/05/2021

Trajectory Prediction using Equivariant Continuous Convolution

Robin Walters, Jinxi Li, Rose Yu

Keywords Paper

argoverse, continuous convolution, trajectory prediction, equivariant, symmetry

0

0

0

0

5:09

18/07/2021

Enhancing Robustness of Neural Networks through Fourier Stabilization

Netanel Raviv, Aidan Kelley, Minzhe Guo, Yevgeniy Vorobeychik

Keywords Paper

Probabilistic Methods, Variational Inference, Algorithms, Boosting and Ensemble Methods; Probabilistic Methods; Probabilistic Methods, Bayesian Theory, Social Aspects of Machine Learning, Privacy, Anonymity, and Security

0

0

0

0

4:57

14/06/2020

Towards Achieving Adversarial Robustness by Enforcing Feature Consistency Across Bit Planes

Sravanti Addepalli, Vivek B.S., Arya Baburaj and
Gaurang Sriramanan, R. Venkatesh Babu

Keywords Paper

adversarial robustness, adversarial defense, adversarial training, fast adversarial training, adversary-free training, adversarial attacks, efficient adversarial training, generalization, feature consistency, deep neural networks

0

0

0

0

1:01

05/01/2021

Improving Robustness and Uncertainty Modelling in Neural Ordinary Differential Equations

Srinivas Anumasa, P. K. Srijith

Keywords Paper

0

0

0

0

4:53

06/12/2021

Generalized Depthwise-Separable Convolutions for Adversarially Robust and Efficient Neural Networks

Hassan Dbouk, Naresh Shanbhag

Keywords Paper

deep learning, robustness, adversarial robustness and security

0

0

0

0

15:01

18/07/2021

Density Constrained Reinforcement Learning

Zengyi Qin, Yuxiao Chen, Chuchu Fan

Keywords Paper

Reinforcement Learning and Planning

0

0

0

0

4:50

06/12/2020

Enabling certification of verification-agnostic networks via memory-efficient semidefinite programming

Sumanth Dathathri, Krishnamurthy Dvijotham, Alexey Kurakin and
Aditi Raghunathan, Jonathan Uesato, Rudy Bunel, Shreya Shankar, Jacob Steinhardt, Ian Goodfellow, Percy Liang, Pushmeet Kohli

Keywords Paper

0

0

0

0

3:23

06/12/2021

Encoding Robustness to Image Style via Adversarial Feature Perturbations

Manli Shu, Zuxuan Wu, Micah Goldblum, Tom Goldstein

Keywords Paper

deep learning, machine learning, robustness, adversarial robustness and security, domain adaptation

0

0

0

0

7:36

04/07/2020

SEEK: Segmented Embedding of Knowledge Graphs

Wentao Xu, Shun Zheng, Liang He and
Bin Shao, Jian Yin, Tie-Yan Liu

Keywords Paper

Segmented Graphs, knowledge embedding, artificial intelligence, recommendation

0

0

0

0

12:01