In this paper, a multi-objective particle swarm optimization algorithm (DEMOPSO) is proposed, which introduces a double elite selection mechanism to select high-quality elite particles in the archive and enhances the ...
详细信息
ISBN:
(数字)9798331520298
ISBN:
(纸本)9798331520304
In this paper, a multi-objective particle swarm optimization algorithm (DEMOPSO) is proposed, which introduces a double elite selection mechanism to select high-quality elite particles in the archive and enhances the convergence and diversity of the population. In addition, the algorithm adjusts the degree of variation according to the particle crowding, which further enhances the diversity of the population and avoids the dilemma of local optimal solutions. Finally, by comparing and analyzing the results with the selected six classical comparison algorithms on the ZDT series of test functions, the experimental results verify that DEMOPSO performs well in terms of convergence and diversity and is able to achieve better optimization results in complex multi-objective optimization problems, demonstrating its superior performance and advantages.
Although adversarial training proves to be an effective method for training robust neural networks, its application is constrained by model capacity, making it challenging to implement in scenarios with limited comput...
详细信息
ISBN:
(数字)9798331520298
ISBN:
(纸本)9798331520304
Although adversarial training proves to be an effective method for training robust neural networks, its application is constrained by model capacity, making it challenging to implement in scenarios with limited computing and storage resources. To tackle this issue, adversarial distillation emerges as a viable solution to transfer knowledge from a robust teacher model to a smaller-capacity student model. However, most existing research overlooks the importance of inconsistencies in adversarial example features during distillation, which can potentially lead to a degradation in the student model's performance. In this paper, we propose a novel method named importance-aware intermediate distillation method (IID), which enhances the efficiency of distillation by leveraging the teacher model's prior knowledge to perceive the varying importance of features in different samples. Through extensive experiments on different datasets, we demonstrate the effectiveness of IID in improving model robustness.
Artificial Intelligence (AI) models use statistical learning over data to solve complex problems for which straightforward rules or algorithms may be difficult or impossible to design; however, a side effect is that m...
详细信息
ISBN:
(数字)9798331534677
ISBN:
(纸本)9798331534684
Artificial Intelligence (AI) models use statistical learning over data to solve complex problems for which straightforward rules or algorithms may be difficult or impossible to design; however, a side effect is that models that are complex enough to sufficiently represent the function may be uninterpretable. Combinatorial testing, a black-box approach arising from software testing, has been applied to test AI models. A key differentiator between traditional software and AI is that many traditional software faults are deterministic, requiring a failureinducing combination of inputs to appear only once in the test set for it to be discovered. On the other hand, AI models learn statistically by reinforcing weights through repeated appearances in the training dataset, and the frequency of input combinations plays a significant role in influencing the model’s behavior. Thus, a single occurrence of a combination of feature values may not be sufficient to influence the model’s behavior. Consequently, measures like simple combinatorial coverage that are applicable to software testing do not capture the frequency with which interactions are covered in the AI model’s input space. This work develops methods to characterize the data frequency coverage of feature interactions in training datasets and analyze the impact of imbalance, or skew, in the combinatorial frequency coverage of the training data on model performance. We demonstrate our methods with experiments on an open-source dataset using several classical machine learning algorithms. This pilot study makes three observations: performance may increase or decrease with data skew, feature importance methods do not predict skew impact, and adding more data may not mitigate skew effects.
Today LoRa data transmission technology is gaining more and more popularity among Internet devices for collecting information and communication. The main purpose of LoRa technology is to collect simple information in ...
详细信息
ISBN:
(数字)9798331532635
ISBN:
(纸本)9798331532642
Today LoRa data transmission technology is gaining more and more popularity among Internet devices for collecting information and communication. The main purpose of LoRa technology is to collect simple information in small packets transmitted at a low speed. Information is collected from various sensors and devices located within the range of the main receiver. This technology complements Wi-Fi. It is suitable for applications with a long range, low transmission speed and power consumption. WiFi provides a short range, low speed and power consumption. LoRa technology is constantly being improved, in connection with which monitoring radio networks built on its basis is of current importance for implementation and study (during the educational process). Modeling of radio communication algorithms in LoRa data transmission technology is one of the important research areas for the development of radio networks of this type. In this regard, this paper develops software that generates files of digital readings of the LoRa radio signal quadratures.
This work presents an automated Resume Analyzer system to streamline the initial resume screening process. Designed to alleviate recruiters’ workloads, the system extracts key resume details—such as name, contact in...
详细信息
ISBN:
(数字)9798331508685
ISBN:
(纸本)9798331519476
This work presents an automated Resume Analyzer system to streamline the initial resume screening process. Designed to alleviate recruiters’ workloads, the system extracts key resume details—such as name, contact information, experience, and qualifications—without manual input. This system leverages Natural Language Processing (NLP) and text mining. A novel scoring mechanism evaluates resumes based on the presence of essential elements like declarations, interests, and skill sets. This web tool provides a resume score which is used by organization and applicants to review about it. This tool allows recruiters to assess resumes in PDF with a rating rapidly, saving them both time and effort. The SVM algorithm outperformed kNN, Naïve Bayes, Decision Tree and Random Forest.
In the current technological landscape, the rise of Web 3.0 is notable. Alongside this, immersive technologies like Virtual Reality (VR) and Augmented Reality (AR) are gaining increased popularity. This is due to idea...
详细信息
ISBN:
(数字)9798331523893
ISBN:
(纸本)9798331523909
In the current technological landscape, the rise of Web 3.0 is notable. Alongside this, immersive technologies like Virtual Reality (VR) and Augmented Reality (AR) are gaining increased popularity. This is due to idea like the Metaverse, which is a spatial computing platform that offers virtual experiences replicating real environments. However, Metaverse also faces security vulnerabilities, such as the risk of identity theft, financial fraud, and the damage of virtual assets due to the public openness of the platform. To mitigate these paramount concerns, there has to be a robust user authentication and verification approach before granting access to these domains. We propose a Convolutional Neural Networks (CNN)-based approach, i.e., MetaSec, which focuses on analyzing incoming user traffic patterns to identify any anomaly in the Internet of Things (IoT)-enabled Metaverse environment. The performance evaluation of MetaSec has been done using various optimization algorithms such as Stochastic Gradient Descent (SGD), Adam, RMSprop, and Adagrad. The main goal of rigorous assessment is to identify the most effective optimization approach for enhancing virtual network security. Thus, MetaSec proactively uncovers potential susceptibility and minimises the risk of intrusion and harm to critical resources. These hardware and software plug-in ranges from computer systems, networks, databases, websites, and other network-based applications and services applied in the virtual domain.
Although Ethereum stands as the dominant blockchain for smart contracts and decentralized applications, faces persistent security challenges from fraudulent activities. Such activities often correlate with off-chain p...
详细信息
ISBN:
(数字)9798331520298
ISBN:
(纸本)9798331520304
Although Ethereum stands as the dominant blockchain for smart contracts and decentralized applications, faces persistent security challenges from fraudulent activities. Such activities often correlate with off-chain platform, such as blog platform and social media. Existing methods analyze fraud activities primarily rely on on-chain transaction data, neglecting interdependencies between on-chain and off-chain activities. In this paper, we observe that there are associations between airdrop campaigns in X platform, a famous social platform and Ethereum fraudulent activities. Further, we crawl Ethereum addresses and posts of these users in airdrop campaigns, and construct a cross-platform datasets from X to Ethereum, including matching pairs of Ethereum addresses to X users, Ethereum transactions and X post data. Due to inherent heterogeneity between Ethereum transactions (structured graphs) and X data (unstructured text/images), we design a multimodal fusion framework leveraging transformer architectures to fuse on-chain transaction features with off-chain content features (text and image representations). Finally, the fused features are leveraged to construct downstream fraud transactions classifiers. Experimental results demonstrate that classifiers using fused features outperform classifiers using transaction features, achieving a 12% improvement in Recall. Our findings highlight the critical role of off-chain data in enhancing fraud detection accuracy.
This article is devoted to the development of a microprocessor-based portable spectrophotometer for the analysis of substances in water, the development of an algorithm for analyzing data from the device, and the writ...
详细信息
ISBN:
(数字)9798331511241
ISBN:
(纸本)9798331511258
This article is devoted to the development of a microprocessor-based portable spectrophotometer for the analysis of substances in water, the development of an algorithm for analyzing data from the device, and the writing of software based on this algorithm. In addition, the article is devoted to assessing the applicability and relevance of this device for high-quality water monitoring. The article presents a structural diagram and an optical system diagram of a microprocessor-based portable spectrophotometer. In addition to the microcontroller, the portable device also uses at least two logical modules: a servo drive and a Bluetooth module. The principle of operation of the device's microcontroller is presented in the form of a structural diagram, the entire developed algorithm is divided into two main stages - a preparatory cycle and the main cycle in which samples are studied. A special algorithm has been developed that allows remote control of the spectrophotometer. In addition, the algorithm for intelligent data analysis has several operating modes and a calibration mode.
As large language models (LLMs) become increasingly embedded in diverse applications, from natural language processing to cybersecurity, the demand for robust privacy-preserving solutions has surged. This paper discus...
详细信息
ISBN:
(数字)9798331520298
ISBN:
(纸本)9798331520304
As large language models (LLMs) become increasingly embedded in diverse applications, from natural language processing to cybersecurity, the demand for robust privacy-preserving solutions has surged. This paper discusses privacy vulnerabilities in LLMs, identifying risks at system, application, and network levels. We categorize and evaluate current privacy-preserving methods, based on homomorphic encryption (HE), trusted execution environments (TEE), and secure multi-party computation (MPC), to assess their effectiveness in mitigating data exposure while supporting LLM performance. Building on these insights, we propose a novel privacy-preserving framework for fine-tuning and inference in LLMs. Our findings highlight existing gaps and propose future directions for developing secure, efficient, and scalable privacy-preserving LLM architectures.
Traceability is important in the software development life cycle for managing the connections among various software artifacts, particularly pull requests (PRs) and issues. Developers often neglect to link these manua...
详细信息
ISBN:
(数字)9798331535100
ISBN:
(纸本)9798331535117
Traceability is important in the software development life cycle for managing the connections among various software artifacts, particularly pull requests (PRs) and issues. Developers often neglect to link these manually, reducing traceability. Existing algorithms to recover these links have limited application in closed-source projects. In this study, we share the experience of using ReLink, a predictive PR-issue linking tool with visualization capabilities, in a closed-source project environment. ReLink stands out due to its availability as a fully functional web application and its semi-automated nature, which enhances usability. ReLink determines missing links between PRs and issues based on a confidence score normalized between 0 and 100, calculated using text similarity and heuristic rules. The tool's effectiveness was evaluated in both an open-source project and an industry-based case study. In the open-source project, ReLink achieved a top-5 accuracy of 0.80 and a precision of 0.77. In the industrial case study, ReLink's effectiveness was validated by practitioners selecting the correct link from five issue suggestions, resulting in a top-5 accuracy of 0.86 and a mean reciprocal rank (MRR) of 0.84. Through this experience, we offer insights into both the benefits and challenges of implementing traceability link recovery in the industry and provide recommendations for practitioners seeking to bridge traceability gaps efficiently.
暂无评论