In this article, we have proposed Deep Learning techniques for cybersecurity in Unmanned Aerial Vehicles (UAVs). UAVs, also known as drones, have become versatile tools used in many applications. However, UA V cyberse...
详细信息
Recently it was shown that the response time of First-Come- First-Served (FCFS) scheduling can be stochastically and asymptotically improved upon by the Nudge scheduling algorithm in case of light-tailed job size dist...
详细信息
Recently, flash-based solid state drives (SSDs) have become a primary storage solution due to their advantages over hard-disk drives. Nonetheless, SSD management presents unique challenges. First, SSDs update data by ...
详细信息
Answer set programming (ASP) is a declarative programming language suited to solve complex combinatorial search problems. Prioritized ASP is the subdiscipline of ASP which aims at prioritizing the models (answer sets)...
详细信息
Evasion attacks on cyber-enabled machine learning (ML) models have recently gained significant traction for their ability to swiftly compel ML models to deviate from their original decisions without substantially affe...
详细信息
Evasion attacks on cyber-enabled machine learning (ML) models have recently gained significant traction for their ability to swiftly compel ML models to deviate from their original decisions without substantially affecting model accuracy during the testing phase. In this article, we initially present a meticulously formulated theoretical framework for a novel and potent evasion attack, leveraging mean-shift perturbation. This attack demonstrates remarkable efficiency in deceiving a wide array of ML models. Subsequently, the urgency of fortifying against such evasion attacks is underscored. It’s worth noting that existing defenses are predominantly model-driven, and their efficacy diminishes when concurrently deployed as a universal defense against both poisoning and evasion attacks. Moreover, empirical evidence from various studies suggests that a single defense mechanism falls short in safeguarding learning models against the myriad forms of adversarial attacks. To alleviate these challenges, we introduced Adaptive Ensemble of Filters (AEF), a defense framework characterized by its robust, transferable, model-agnostic, input distribution-independent, and cross-model-supportive nature. The AEF strategically selects filters to safeguard a target ML model from various well-known poisoning (e.g., Metapoison) and evasion (utilizing mean-shift perturbations, JSMA, FGSM, PGD, BIM, and C&W) attacks, establishing itself as a universal defense against diverse adversarial attacks. Theoretical analysis assures the existence of optimal filter ensembles across different input distributions and adversarial attack landscapes, without encountering mode collapses and vanishing gradients. Our claims are substantiated through validation on three publicly available image datasets—MNIST, CIFAR-10, and EuroSAT. IEEE
Real-world images often encompass embedded texts that adhere to disparate disciplines like business, education, and amusement, to name a few. Such images are graphically rich in terms of font attributes, color distrib...
详细信息
Sorting networks are sorting algorithms that execute a sequence of operations independently of the input. Since they can be implemented directly as circuits, sorting networks are easy to implement in hardware – but t...
详细信息
In this paper, we present the preliminary experiments for the development of an ingestion mechanism to move data from Electronic Health Records to machine learning processes, based on the concept of Linked Data and th...
详细信息
To address the premature convergence and search stagnation of arithmetic optimization algorithm (AOA), the paper proposes a hybrid arithmetic optimization algorithm (HAOA) and applies it to the practical robot path pl...
详细信息
We propose a novel way of combining two canonical models used in Artificial Intelligence (AI): Bayesian Networks (BN) and Ant Colony Optimisation (ACO) in order to obtain a fast graph-traversal algorithm that establis...
详细信息
暂无评论