ICLR 2018 accepted papers on Adversarial robustness and security
21 of the 337 papers accepted at ICLR 2018 (6.2%) are on Adversarial robustness and security, compared with 2.5% at ICLR 2017. They fall into 3 topics.
See ICLR 2018 in the interactive map
Submitting to ICLR? Checklist →
Working on something in this area? Describe your idea in scime atlas to see the closest papers by meaning and how crowded the spot has become.
Topics and papers
Certified and empirical robustness · 12 papers
- mixup: Beyond Empirical Risk Minimization (4,723 citations)
- Towards Deep Learning Models Resistant to Adversarial Attacks (1,508 citations)
- Certified Defenses against Adversarial Examples (558 citations)
- Characterizing Adversarial Subspaces Using Local Intrinsic Dimensionality (486 citations)
- Thermometer Encoding: One Hot Way To Resist Adversarial Examples (455 citations)
- Certifying Some Distributional Robustness with Principled Adversarial Training (326 citations)
- PixelDefend: Leveraging Generative Models to Understand and Defend against Adversarial Examples (285 citations)
- Stochastic Activation Pruning for Robust Adversarial Defense (200 citations)
- Mitigating Adversarial Effects Through Randomization (195 citations)
- Evaluating the Robustness of Neural Networks: An Extreme Value Theory Approach (154 citations)
- Decision Boundary Analysis of Adversarial Examples (86 citations)
- Robustness of Classifiers to Universal Perturbations: A Geometric Perspective (43 citations)
Black-box and transfer attacks · 7 papers
- Ensemble Adversarial Training: Attacks and Defenses (2,314 citations)
- Generating Natural Adversarial Examples (456 citations)
- Countering Adversarial Images using Input Transformations (429 citations)
- Towards Reverse-Engineering Black-Box Neural Networks (269 citations)
- Decision-Based Adversarial Attacks: Reliable Attacks Against Black-Box Machine Learning Models (133 citations)
- Cascade Adversarial Machine Learning Regularized with a Unified Embedding (62 citations)
- Attacking Binarized Neural Networks (22 citations)
