| When does machine learning {FAIL}? generalized transferability for evasion and poisoning attacks O Suciu, R Marginean, Y Kaya, H Daume III, T Dumitras 27th USENIX Security Symposium (USENIX Security 18), 1299-1316, 2018 | 357 | 2018 |
| Shallow-Deep Networks: Understanding and Mitigating Network Overthinking Y Kaya, S Hong, T Dumitras International Conference on Machine Learning, 3301-3310, 2019 | 333 | 2019 |
| Terminal Brain Damage: Exposing the Graceless Degradation in Deep Neural Networks Under Hardware Fault Attacks S Hong, P Frigo, Y Kaya, C Giuffrida, T Dumitraş 28th USENIX Security Symposium (USENIX Security 19), 497-514, 2019 | 247 | 2019 |
| On the effectiveness of mitigating data poisoning attacks with gradient shaping S Hong, V Chandrasekaran, Y Kaya, T Dumitraş, N Papernot arXiv preprint arXiv:2002.11497, 2020 | 154 | 2020 |
| Security analysis of deep neural networks operating in the presence of cache side-channel attacks S Hong, M Davinroy, Y Kaya, SN Locke, I Rackow, K Kulda, ... arXiv preprint arXiv:1810.03487, 2018 | 97 | 2018 |
| When does data augmentation help with membership inference attacks? Y Kaya, T Dumitras International conference on machine learning, 5345-5355, 2021 | 73 | 2021 |
| A Panda? No, It's a Sloth: Slowdown Attacks on Adaptive Multi-Exit Neural Network Inference S Hong, Y Kaya, IV Modoranu, T Dumitras International Conference on Learning Representations 2021, 2021 | 69 | 2021 |
| How to 0wn NAS in your spare time S Hong, M Davinroy, Y Kaya, D Dachman-Soled, T Dumitraş International Conference on Learning Representations 2020, 2020 | 42 | 2020 |
| On the effectiveness of regularization against membership inference attacks Y Kaya, S Hong, T Dumitras arXiv preprint arXiv:2006.05336, 2020 | 36 | 2020 |
| Qu-anti-zation: Exploiting quantization artifacts for achieving adversarial outcomes S Hong, MA Panaitescu-Liess, Y Kaya, T Dumitras Advances in Neural Information Processing Systems 34, 9303-9316, 2021 | 19 | 2021 |
| Security analysis of deep neural networks operating in the presence of cache side-channel attacks. CoRR abs/1810.03487 (2018) S Hong, M Davinroy, Y Kaya, SN Locke, I Rackow, K Kulda, ... arXiv preprint arXiv:1810.03487, 2018 | 8 | 2018 |
| Generating distributional adversarial examples to evade statistical detectors Y Kaya, B Zafar, S Aydore, N Rauschmayr, K Kenthapadi | 6 | 2022 |
| Demystifying Behavior-Based Malware Detection at Endpoints Y Kaya, Y Chen, S Saha, F Pierazzi, L Cavallaro, D Wagner, T Dumitras arXiv preprint arXiv:2405.06124, 2024 | 5 | 2024 |
| DRSM: De-Randomized Smoothing on Malware Classifier Providing Certified Robustness S Saha, W Wang, Y Kaya, S Feizi, T Dumitras The Twelfth International Conference on Learning Representations (ICLR 2024), 2024 | 2 | 2024 |
| The Limitations of Deep Learning Methods in Realistic Adversarial Settings Y Kaya | 1 | 2023 |
| Understanding, Uncovering, and Mitigating the Causes of Inference Slowdown for Language Models K Varma, A Numanoğlu, Y Kaya, T Dumitraş 2024 IEEE Conference on Secure and Trustworthy Machine Learning (SaTML), 723-740, 2024 | | 2024 |
| Like Oil and Water: Group Robustness Methods and Poisoning Defenses Don’t Mix MA Panaitescu-Liess, Y Kaya, S Zhu, F Huang, T Dumitras The Twelfth International Conference on Learning Representations, 2024 | | 2024 |
| Too Big to FAIL: What You Need to Know Before Attacking a Machine Learning System T Dumitraş, Y Kaya, R Mărginean, O Suciu Security Protocols XXVI: 26th International Workshop, Cambridge, UK, March …, 2018 | | 2018 |
| PoisonedParrot: Subtle Data Poisoning Attacks to Elicit Copyright-Infringing Content from Large Language Models MA Panaitescu-Liess, P Pathmanathan, Y Kaya, Z Che, B An, S Zhu, ... Neurips Safe Generative AI Workshop 2024, 0 | | |
Tudor DumitrasAssistant Professor, University of Maryland, College ParkVerified email at umiacs.umd.edu