More Like this

1. Tsirelson Polytopes and randomness generation

   https://doi.org/10.1088/1367-2630/aba30d  

   Bierhorst, Peter; Zhang, Yanbao (August 2020, New Journal of Physics)

   Abstract We classify the extreme points of a polytope of probability distributions in the (2,2,2) CHSH-Bell setting that is induced by a single Tsirelson bound. We do the same for a class of polytopes obtained from a parametrized family of multiple Tsirelson bounds interacting non-trivially. Such constructions can be applied to device-independent random number generation using the method of probability estimation factors (2018 Phys. Rev. A98040304(R)). We demonstrate a meaningful improvement in certified randomness applying the new polytopes characterized here. 

   more » « less
2. Linear Optical Logical Bell State Measurements with Optimal Loss-Tolerance Threshold

   https://doi.org/10.1103/PRXQuantum.4.040322  

   Hilaire, Paul; Castor, Yaron; Barnes, Edwin; Economou, Sophia E; Grosshans, Frédéric (November 2023, PRX Quantum)

   Quantum threshold theorems impose hard limits on the hardware capabilities to process quantum information. We derive tight and fundamental upper bounds to loss-tolerance thresholds in different linear-optical quantum information processing settings through an adversarial framework, taking into account the intrinsically probabilistic nature of linear optical Bell measurements. For logical Bell state measurements—ubiquitous operations in photonic quantum information—we demonstrate analytically that linear optics can achieve the fundamental loss threshold imposed by the no-cloning theorem even though, following the work of Lee et al. [Phys. Rev. A 100, 052303 (2019)] the constraint was widely assumed to be stricter. We spotlight the assumptions of the latter publication and find their bound holds for a logical Bell measurement built from adaptive physical linear-optical Bell measurements. We also give an explicit even stricter bound for nonadaptive Bell measurements. 

   more » « less
3. Fast Confidence Detection: One Hot Way to Detect Adversarial Attacks via Sensor Pattern Noise Fingerprinting

   https://doi.org/10.1145/3289602.3293975  

   Lan, Yazhu; Guo, Qingli; Zhang, Guohe; Xu, Yuanchao; Nixon, Kent W.; Li, Hai Helen; Chen, Yiran (February 2019, International Symposium on Field-Programmable Gate Arrays)

   Deep Neural Networks (DNNs) have shown phenomenal success in a wide range of real-world applications. However, a concerning weakness of DNNs is that they are vulnerable to adversarial attacks. Although there exist methods to detect adversarial attacks, they often suffer constraints on specific attack types and provide limited information to downstream systems. We specifically note that existing adversarial detectors are often binary classifiers, which differentiate clean or adversarial examples. However, detection of adversarial examples is much more complicated than such a scenario. Our key insight is that the confidence probability of detecting an input sample as an adversarial example will be more useful for the system to properly take action to resist potential attacks. In this work, we propose an innovative method for fast confidence detection of adversarial attacks based on integrity of sensor pattern noise embedded in input examples. Experimental results show that our proposed method is capable of providing a confidence distribution model of most of popular adversarial attacks. Furthermore, our presented method can provide early attack warning with even the attack types based on different properties of the confidence distribution models. Since fast confidence detection is a computationally heavy task, we propose an FPGA-Based hardware architecture based on a series of optimization techniques, such as incremental multi-level quantization and etc. We realize our proposed method on an FPGA platform and achieve a high efficiency of 29.740 IPS/W with a power consumption of only 0.7626W. 

   more » « less
4. Learning Robust and Privacy-Preserving Representations via Information Theory

   https://doi.org/10.1609/AAAI.V39I21.34392  

   Zhang, Binghui; Noorbakhsh, Sayedeh Leila; Dong, Yun; Hong, Yuan; Wang, Binghui (April 2025, Proceedings of the AAAI Conference on Artificial Intelligence)

   Machine learning models are vulnerable to both security attacks (e.g., adversarial examples) and privacy attacks (e.g., private attribute inference). We take the first step to mitigate both the security and privacy attacks, and maintain task utility as well. Particularly, we propose an information-theoretic framework to achieve the goals through the lens of representation learning, i.e., learning representations that are robust to both adversarial examples and attribute inference adversaries. We also derive novel theoretical results under our framework, e.g., the inherent trade-off between adversarial robustness/utility and attribute privacy, and guaranteed attribute privacy leakage against attribute inference adversaries. 

   more » « less
5. Keeping up with dynamic attackers: Certifying robustness to adaptive online data poisoning

   Bose, Avinandan; Lessard, Laurent; Fazel, Maryam; Dvijotham, Krishnamurthy (April 2025, Proceedings of Machine Learning Research)

   The rise of foundation models fine-tuned on human feedback from potentially untrusted users has increased the risk of adversarial data poisoning, necessitating the study of robustness of learning algorithms against such attacks. Existing research on provable certified robustness against data poisoning attacks primarily focuses on certifying robustness for static adversaries who modify a fraction of the dataset used to train the model before the training algorithm is applied. In practice, particularly when learning from human feedback in an online sense, adversaries can observe and react to the learning process and inject poisoned samples that optimize adversarial objectives better than when they are restricted to poisoning a static dataset once, before the learning algorithm is applied. Indeed, it has been shown in prior work that online dynamic adversaries can be significantly more powerful than static ones. We present a novel framework for computing certified bounds on the impact of dynamic poisoning, and use these certificates to design robust learning algorithms. We give an illustration of the framework for the mean estimation problem and binary classification problems and outline directions for extending this in further work. 

   more » « less