PhD student at Max Planck Institute for Informatics; working on adversarial robustness; blog davidstutz.de.

Second-Order Adversarial Attack and Certifiable Robustness

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

Certified Robustness to Adversarial Examples with Differential Privacy

arXiv e-Print archive - 2018 via Local arXiv

Keywords: stat.ML, cs.AI, cs.CR, cs.LG

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: stat.ML, cs.AI, cs.CR, cs.LG

ImageNet-trained {CNN}s are biased towards texture; increasing shape bias improves accuracy and robustness

International Conference on Learning Representations - 2019 via Local Bibsonomy

Keywords: deep-learning, machine-learning, stable, foundations, robustness, theory

International Conference on Learning Representations - 2019 via Local Bibsonomy

Keywords: deep-learning, machine-learning, stable, foundations, robustness, theory

Approximating CNNs with Bag-of-local-Features models works surprisingly well on ImageNet

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

Towards Stable and Efficient Training of Verifiably Robust Neural Networks

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

Efficient Neural Network Robustness Certification with General Activation Functions

Neural Information Processing Systems Conference - 2018 via Local Bibsonomy

Keywords: dblp

Neural Information Processing Systems Conference - 2018 via Local Bibsonomy

Keywords: dblp

Generalization in Deep Networks: The Role of Distance from Initialization

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

On the Effectiveness of Interval Bound Propagation for Training Verifiably Robust Models

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CR, stat.ML

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CR, stat.ML

Batch Normalization is a Cause of Adversarial Vulnerability

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

Radial basis function neural networks: a topical state-of-the-artsurvey

Open Computer Science - 2016 via Local Bibsonomy

Keywords: dblp

Open Computer Science - 2016 via Local Bibsonomy

Keywords: dblp

How Can We Be So Dense? The Benefits of Using Highly Sparse Representations

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

Deep-RBF Networks Revisited: Robust Classification with Rejection

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

Neural Networks with Structural Resistance to Adversarial Attacks

arXiv e-Print archive - 2018 via Local arXiv

Keywords: stat.ML, cs.CR, cs.LG, cs.NE

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: stat.ML, cs.CR, cs.LG, cs.NE

Adversarial Examples Are Not Bugs, They Are Features

- 2019 via Local Bibsonomy

Keywords: adversarial

- 2019 via Local Bibsonomy

Keywords: adversarial

Bit-Flip Attack: Crushing Neural Network withProgressive Bit Search

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

The Lottery Ticket Hypothesis: Finding Sparse, Trainable Neural Networks

arXiv e-Print archive - 2019 via Local arXiv

Keywords: cs.LG, cs.AI, cs.NE

more

arXiv e-Print archive - 2019 via Local arXiv

Keywords: cs.LG, cs.AI, cs.NE

Certified Adversarial Robustness via Randomized Smoothing

arXiv e-Print archive - 2019 via Local arXiv

Keywords: cs.LG, stat.ML

more

arXiv e-Print archive - 2019 via Local arXiv

Keywords: cs.LG, stat.ML

Enhancing The Reliability of Out-of-distribution Image Detection in Neural Networks

arXiv e-Print archive - 2017 via Local arXiv

Keywords: cs.LG, stat.ML

more

arXiv e-Print archive - 2017 via Local arXiv

Keywords: cs.LG, stat.ML

Adding Gradient Noise Improves Learning for Very Deep Networks

arXiv e-Print archive - 2015 via Local arXiv

Keywords: stat.ML, cs.LG

more

arXiv e-Print archive - 2015 via Local arXiv

Keywords: stat.ML, cs.LG

Training Confidence-calibrated Classifiers for Detecting Out-of-Distribution Samples

International Conference on Learning Representations - 2018 via Local Bibsonomy

Keywords: dblp

International Conference on Learning Representations - 2018 via Local Bibsonomy

Keywords: dblp

The Limitations of Adversarial Training and the Blind-Spot Attack

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

A Theoretical Framework for Robustness of (Deep) Classifiers against Adversarial Examples

arXiv e-Print archive - 2016 via Local arXiv

Keywords: cs.LG, cs.CR, cs.CV

more

arXiv e-Print archive - 2016 via Local arXiv

Keywords: cs.LG, cs.CR, cs.CV

Towards Poisoning of Deep Learning Algorithms with Back-gradient Optimization

Proceedings of the 10th ACM Workshop on Artificial Intelligence and Security - AISec '17 - 2017 via Local CrossRef

Keywords:

Proceedings of the 10th ACM Workshop on Artificial Intelligence and Security - AISec '17 - 2017 via Local CrossRef

Keywords:

MagNet: A Two-Pronged Defense against Adversarial Examples

ACM ACM Conference on Computer and Communications Security - 2017 via Local Bibsonomy

Keywords: dblp

ACM ACM Conference on Computer and Communications Security - 2017 via Local Bibsonomy

Keywords: dblp

UPSET and ANGRI : Breaking High Performance Image Classifiers

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

On the importance of single directions for generalization

arXiv e-Print archive - 2018 via Local arXiv

Keywords: stat.ML, cs.AI, cs.LG, cs.NE

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: stat.ML, cs.AI, cs.LG, cs.NE

Improving Transferability of Adversarial Examples with Input Diversity

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.CV, cs.LG, stat.ML

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.CV, cs.LG, stat.ML

Improving Network Robustness against Adversarial Attacks with Compact Convolution

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

Regularizing Neural Networks by Penalizing Confident Output Distributions

arXiv e-Print archive - 2017 via Local arXiv

Keywords: cs.NE, cs.LG

more

arXiv e-Print archive - 2017 via Local arXiv

Keywords: cs.NE, cs.LG

Beyond Pixel Norm-Balls: Parametric Adversaries using an Analytically Differentiable Renderer

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CV, cs.GR, stat.ML

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CV, cs.GR, stat.ML

Enhanced Attacks on Defensively Distilled Deep Neural Networks

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

Breaking Transferability of Adversarial Samples with Randomness

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

Cost-Sensitive Robustness against Adversarial Examples

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

Interpretability Beyond Feature Attribution: Quantitative Testing with Concept Activation Vectors (TCAV)

arXiv e-Print archive - 2017 via Local arXiv

Keywords: stat.ML

more

arXiv e-Print archive - 2017 via Local arXiv

Keywords: stat.ML

Black-box Adversarial Attacks with Limited Queries and Information

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.CV, cs.CR, stat.ML

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.CV, cs.CR, stat.ML

On the Intriguing Connections of Regularization, Input Gradients and Transferability of Evasion and Poisoning Attacks

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CR, stat.ML, 68T10, 68T45

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CR, stat.ML, 68T10, 68T45

Attacks Meet Interpretability: Attribute-steered Detection of Adversarial Samples

Neural Information Processing Systems Conference - 2018 via Local Bibsonomy

Keywords: dblp

Neural Information Processing Systems Conference - 2018 via Local Bibsonomy

Keywords: dblp

Adversarial Dropout for Supervised and Semi-Supervised Learning

AAAI Conference on Artificial Intelligence - 2018 via Local Bibsonomy

Keywords: dblp

AAAI Conference on Artificial Intelligence - 2018 via Local Bibsonomy

Keywords: dblp

Fine-Pruning: Defending Against Backdooring Attacks on Deep Neural Networks

Springer RAID - 2018 via Local Bibsonomy

Keywords: dblp

Springer RAID - 2018 via Local Bibsonomy

Keywords: dblp

On the Geometry of Adversarial Examples

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

The Limitations of Model Uncertainty in Adversarial Settings

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

Towards Interpretable Deep Neural Networks by Leveraging Adversarial Examples

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

The Secret Sharer: Measuring Unintended Neural Network Memorization & Extracting Secrets

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

Mitigating Evasion Attacks to Deep Neural Networks via Region-based Classification

Proceedings of the 33rd Annual Computer Security Applications Conference on - ACSAC 2017 - 2017 via Local CrossRef

Keywords:

Proceedings of the 33rd Annual Computer Security Applications Conference on - ACSAC 2017 - 2017 via Local CrossRef

Keywords:

Curriculum Adversarial Training

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence - 2018 via Local CrossRef

Keywords:

Proceedings of the Twenty-Seventh International Joint Conference on Artificial Intelligence - 2018 via Local CrossRef

Keywords:

AI2: Safety and Robustness Certification of Neural Networks with Abstract Interpretation

IEEE Computer Society IEEE Symposium on Security and Privacy - 2018 via Local Bibsonomy

Keywords: dblp

IEEE Computer Society IEEE Symposium on Security and Privacy - 2018 via Local Bibsonomy

Keywords: dblp

Towards Robust Interpretability with Self-Explaining Neural Networks

Neural Information Processing Systems Conference - 2018 via Local Bibsonomy

Keywords: dblp

Neural Information Processing Systems Conference - 2018 via Local Bibsonomy

Keywords: dblp

Efficient Repair of Polluted Machine Learning Systems via Causal Unlearning

Proceedings of the 2018 on Asia Conference on Computer and Communications Security - ASIACCS '18 - 2018 via Local CrossRef

Keywords:

Proceedings of the 2018 on Asia Conference on Computer and Communications Security - ASIACCS '18 - 2018 via Local CrossRef

Keywords:

SoK: Science, Security and the Elusive Goal of Security as a Scientific Pursuit

IEEE Computer Society IEEE Symposium on Security and Privacy - 2017 via Local Bibsonomy

Keywords: dblp

IEEE Computer Society IEEE Symposium on Security and Privacy - 2017 via Local Bibsonomy

Keywords: dblp

Model-Reuse Attacks on Deep Learning Systems

Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security - CCS '18 - 2018 via Local CrossRef

Keywords:

Proceedings of the 2018 ACM SIGSAC Conference on Computer and Communications Security - CCS '18 - 2018 via Local CrossRef

Keywords:

Playing the Game of Universal Adversarial Perturbations

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CV, stat.ML

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CV, stat.ML

Secure Kernel Machines against Evasion Attacks

Proceedings of the 2016 ACM Workshop on Artificial Intelligence and Security - ALSec '16 - 2016 via Local CrossRef

Keywords:

Proceedings of the 2016 ACM Workshop on Artificial Intelligence and Security - ALSec '16 - 2016 via Local CrossRef

Keywords:

Are adversarial examples inevitable?

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

On the Robustness of Convolutional Neural Networks to Internal Architecture and Weight Perturbations

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

Adversarial Initialization - when your network performs the way I want

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2019 via Local Bibsonomy

Keywords: dblp

Fault injection attack on deep neural network

2017 IEEE/ACM International Conference on Computer-Aided Design (ICCAD) - 2017 via Local CrossRef

Keywords:

2017 IEEE/ACM International Conference on Computer-Aided Design (ICCAD) - 2017 via Local CrossRef

Keywords:

Robustness of Generalized Learning Vector Quantization Models against Adversarial Attacks

arXiv e-Print archive - 2019 via Local arXiv

Keywords: cs.LG, cs.AI, cs.CV, stat.ML

more

arXiv e-Print archive - 2019 via Local arXiv

Keywords: cs.LG, cs.AI, cs.CV, stat.ML

Protecting Intellectual Property of Deep Neural Networks with Watermarking

ACM AsiaCCS - 2018 via Local Bibsonomy

Keywords: dblp

ACM AsiaCCS - 2018 via Local Bibsonomy

Keywords: dblp

Fortified Networks: Improving the Robustness of Deep Networks by Modeling the Manifold of Hidden Representations

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

Towards the first adversarially robust neural network model on MNIST

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.CV

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.CV

AE-GAN: adversarial eliminating with GAN

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

Constructing Unrestricted Adversarial Examples with Generative Models

Neural Information Processing Systems Conference - 2018 via Local Bibsonomy

Keywords: dblp

Neural Information Processing Systems Conference - 2018 via Local Bibsonomy

Keywords: dblp

Is Robustness the Cost of Accuracy? – A Comprehensive Study on the Robustness of 18 Deep Image Classification Models

Computer Vision – ECCV 2018 - 2018 via Local CrossRef

Keywords:

Computer Vision – ECCV 2018 - 2018 via Local CrossRef

Keywords:

Generating Natural Adversarial Examples

International Conference on Learning Representations - 2018 via Local Bibsonomy

Keywords: dblp

International Conference on Learning Representations - 2018 via Local Bibsonomy

Keywords: dblp

DisturbLabel: Regularizing CNN on the Loss Layer

Conference and Computer Vision and Pattern Recognition - 2016 via Local CrossRef

Keywords:

Conference and Computer Vision and Pattern Recognition - 2016 via Local CrossRef

Keywords:

Defensive Distillation is Not Robust to Adversarial Examples

arXiv e-Print archive - 2016 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2016 via Local Bibsonomy

Keywords: dblp

Adversarial Training Versus Weight Decay

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, stat.ML

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, stat.ML

Adaptive data augmentation for image classification

2016 IEEE International Conference on Image Processing (ICIP) - 2016 via Local CrossRef

Keywords:

2016 IEEE International Conference on Image Processing (ICIP) - 2016 via Local CrossRef

Keywords:

Analyzing the Robustness of Nearest Neighbors to Adversarial Examples

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

On the Suitability of Lp-Norms for Creating and Preventing Adversarial Examples

Conference and Computer Vision and Pattern Recognition - 2018 via Local Bibsonomy

Keywords: dblp

Conference and Computer Vision and Pattern Recognition - 2018 via Local Bibsonomy

Keywords: dblp

Learning to Compose Domain-Specific Transformations for Data Augmentation.

Neural Information Processing Systems Conference - 2017 via Local dblp

Keywords:

Neural Information Processing Systems Conference - 2017 via Local dblp

Keywords:

Deep k-Nearest Neighbors: Towards Confident, Interpretable and Robust Deep Learning

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, stat.ML

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, stat.ML

Towards Imperceptible and Robust Adversarial Example Attacks against Neural Networks

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CR, stat.ML

more

arXiv e-Print archive - 2018 via Local arXiv

Keywords: cs.LG, cs.CR, stat.ML

On Calibration of Modern Neural Networks

arXiv e-Print archive - 2017 via Local arXiv

Keywords: cs.LG

more

arXiv e-Print archive - 2017 via Local arXiv

Keywords: cs.LG

Interpretation of Neural Networks is Fragile

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2017 via Local Bibsonomy

Keywords: dblp

Spatially Transformed Adversarial Examples

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

arXiv e-Print archive - 2018 via Local Bibsonomy

Keywords: dblp

Attacking the Madry Defense Model with $L_1$-based Adversarial Examples

arXiv e-Print archive - 2017 via Local arXiv

Keywords: stat.ML, cs.CR, cs.LG

more

arXiv e-Print archive - 2017 via Local arXiv

Keywords: stat.ML, cs.CR, cs.LG