Related papers: Adversarial Learning Game for Intrusion Detection …

Adversarial training has gained great popularity as one of the most effective defenses for deep neural network and more generally for gradient-based machine learning models against adversarial perturbations on data points. This paper…

In response to the rapidly evolving nature of adversarial attacks against visual classifiers, numerous defenses have been proposed to generalize against as many known attacks as possible. However, designing a defense method that generalizes…

Neural networks are known to be vulnerable to adversarial attacks -- slight but carefully constructed perturbations of the inputs which can drastically impair the network's performance. Many defense methods have been proposed for improving…

LLM-integrated applications and agents are vulnerable to prompt injection attacks, where adversaries embed malicious instructions within seemingly benign input data to manipulate the LLM's intended behavior. Recent defenses based on…

In modern transportation networks, adversaries can manipulate routing algorithms using false data injection attacks, such as simulating heavy traffic with multiple devices running crowdsourced navigation applications, to mislead vehicles…

Machine learning models can learn from data samples to carry out various tasks efficiently. When data samples are adversarially manipulated, such as by insertion of carefully crafted noise, it can cause the model to make mistakes. Quantum…

While deep neural networks have shown impressive performance in many tasks, they are fragile to carefully designed adversarial attacks. We propose a novel adversarial training-based model by Attention Guided Knowledge Distillation and…

Operating in a dynamic real world environment requires a forward thinking and adversarial aware design for classifiers, beyond fitting the model to the training data. In such scenarios, it is necessary to make classifiers - a) harder to…

Decoy state method closes source security loophole in quantum key distribution (QKD) using laser source. In this method, accurate estimates of the detection rates of vacuum and single photon events plus the error rate of single photon…

We simulate quantum key distribution (QKD) experimental setups and give out some improvement for QKD procedures. A new data post-processing protocol is introduced, mainly including error correction and privacy amplification. This protocol…

Quantum key distribution (QKD) offers the promise of absolutely secure communications. However, proofs of absolute security often assume perfect implementation from theory to experiment. Thus, existing systems may be prone to insidious…

Attack detection is usually approached as a classification problem. However, standard classification tools often perform poorly because an adaptive attacker can shape his attacks in response to the algorithm. This has led to the recent…

Semi-device-independent quantum key distribution aims to achieve a balance between the highest level of security, device independence, and experimental feasibility. Semi-quantum key distribution presents an intriguing approach that seeks to…

Machine Learning (ML) techniques can facilitate the automation of malicious software (malware for short) detection, but suffer from evasion attacks. Many studies counter such attacks in heuristic manners, lacking theoretical guarantees and…

Distillation attacks create a deployment trade-off for model providers: the same outputs that make a model more useful can also make it easier to imitate. We study this trade-off through a minimax game between a utility-constrained teacher…

Quantum key distribution (QKD) permits information-theoretically secure transmission of digital encryption keys, assuming that the behaviour of the devices employed for the key exchange can be reliably modelled and predicted. Remarkably, no…

Quantum Key Distribution (QKD) can provide information-theoretically secure communications and is a strong candidate for the next generation of cryptography. However, in practice, the performance of QKD is limited by "practical…

Quantum key distribution (QKD) allows two users to exchange a provably secure key for cryptographic applications. In prepare-and-measure QKD protocols, the states must be indistinguishable to prevent information leakage to an eavesdropper…

Quantum key distribution (QKD offers a long-term solution to establish information-theoretically secure keys between two distant users. In practice, with a careful characterization of quantum sources and the decoy-state method,…

K-Nearest Neighbor (kNN)-based deep learning methods have been applied to many applications due to their simplicity and geometric interpretability. However, the robustness of kNN-based classification models has not been thoroughly explored…