Auflistung nach:
Auflistung it - Information Technology 65(4-5) - November 2023 nach Erscheinungsdatum
1 - 9 von 9
Treffer pro Seite
Sortieroptionen
- ZeitschriftenartikelMachine learning in run-time control of multicore processor systems(it - Information Technology: Vol. 65, No. 4-5, 2023) Maurer, Florian; Thoma, Moritz; Surhonne, Anmol Prakash; Donyanavard, Bryan; Herkersdorf, AndreasModern embedded and cyber-physical applications consist of critical and non-critical tasks co-located on multiprocessor systems on chip (MPSoCs). Co-location of tasks results in contention for shared resources, resulting in interference on interconnect, processing units, storage, etc. Hence, machine learning-based resource managers must operate even non-critical tasks within certain constraints to ensure proper execution of critical tasks. In this paper we demonstrate and evaluate countermeasures based on backup policies to enhance rule-based reinforcement learning to enforce constraints. Detailed experiments reveal the CPUs’ performance degradation caused by different designs, as well as their effectiveness in preventing constraint violations. Further, we exploit the interpretability of our approach to further improve the resource manager’s operation by adding designers’ experience into the rule set.
- ZeitschriftenartikelMachine learning and cyber security(it - Information Technology: Vol. 65, No. 4-5, 2023) Karius, Sebastian; Knöchel, Mandy; Heße, Sascha; Reiprich, TimCyber Security has gained a significant amount of perceived importance when talking about the risks and challenges that lie ahead in the field of information technology. A recent increase in high-profile incidents involving any form of cyber criminality have raised the awareness of threats that were formerly often hidden from public perception, e.g., with openly carried out attacks against critical infrastructure to accompany traditional forms of warfare, extending those to the cyberspace. Add to that very personal experience of everyday social engineering attacks, which are cast out like a fishing net on a large scale, e.g., to catch anyone not careful enough to double-check a suspicious email. But as the threat level rises and the attacks become even more sophisticated, so do the methods to mitigate (or at least recognize) them. Of central importance here are methods from the field of machine learning (ML). This article provides a comprehensive overview of applied ML methods in cyber security, illustrates the importance of ML for cyber security, and discusses issues and methods for generating good datasets for the training phase of ML methods used in cyber security. This includes own work on the topics of network traffic classification, the collection of real-world attacks using honeypot systems as well as the use of ML to generate artificial network traffic.
- ZeitschriftenartikelMachine learning in AI Factories – five theses for developing, managing and maintaining data-driven artificial intelligence at large scale(it - Information Technology: Vol. 65, No. 4-5, 2023) Hildesheim, Wolfgang; Holoyad, Taras; Schmid, ThomasThe use of artificial intelligence (AI) is today’s dominating technological trend across all industries. With the maturing of deep learning and other data-driven techniques, AI has over the last decade become an essential component for an increasing number of products and services. In parallel to this development, technological advances have been accelerating the production of novel AI models from large-scale datasets. This global phenomenon has been driving the need for an efficient industrialized approach to develop, manage and maintain AI models at large scale. Such an approach is provided by the state-of-the-art operational concept termed AI Factory, which refers to an infrastructure for AI models and implements the idea of AI as a Service (AIaaS). Moreover, it ensures performance, transparency and reproducibility of AI models at any point in the continuous AI development process. This concept, however, does not only require new technologies and architectures, but also new job roles. Here, we discuss current trends, outline requirements and identify success factors for AI Factories. We conclude with recommendations for their successful use in practice as well as perspectives on future developments.
- ZeitschriftenartikelMachine learning applications(it - Information Technology: Vol. 65, No. 4-5, 2023) Arndt, Natanael; Molitor, Paul; Usbeck, Ricardo
- ZeitschriftenartikelMachine learning in computational literary studies(it - Information Technology: Vol. 65, No. 4-5, 2023) Hatzel, Hans Ole; Stiemer,Haimo; Biemann, Chris; Gius, EvelynIn this article, we provide an overview of machine learning as it is applied in computational literary studies, the field of computational analysis of literary texts and literature related phenomena. We survey a number of scientific publications for the machine learning methodology the scholars used and explain concepts of machine learning and natural language processing while discussing our findings. We establish that besides transformer-based language models, researchers still make frequent use of more traditional, feature-based machine learning approaches; possible reasons for this are to be found in the challenging application of modern methods to the literature domain and in the more transparent nature of traditional approaches. We shed light on how machine learning-based approaches are integrated into a research process, which often proceeds primarily from the non-quantitative, interpretative approaches of non-digital literary studies. Finally, we conclude that the application of large language models in the computational literary studies domain may simplify the application of machine learning methodology going forward, if adequate approaches for the analysis of literary texts are found.
- ZeitschriftenartikelArtificial intelligence for molecular communication(it - Information Technology: Vol. 65, No. 4-5, 2023) Bartunik, Max; Kirchner, Jens; Keszocze, OliverMolecular communication is a novel approach for data transmission between miniaturised devices, especially in contexts where electrical signals are to be avoided. The communication is based on sending molecules (or other particles) at nanoscale through a typically fluid channel instead of the “classical” approach of sending electrons over a wire. Molecular communication devices have a large potential in future medical applications as they offer an alternative to antenna-based transmission systems that may not be applicable due to size, temperature, or radiation constraints. The communication is achieved by transforming a digital signal into concentrations of molecules that represent the signal. These molecules are then detected at the other end of the communication channel and transformed back into a digital signal. Accurately modeling the transmission channel is often not possible which may be due to a lack of data or time-varying parameters of the channel (e.g., the movements of a person wearing a medical device). This makes the process of demodulating the signal (i.e., signal classification) very difficult. Many approaches for demodulation have been discussed in the literature with one particular approach having tremendous success – artificial neural networks. These artificial networks imitate the decision process in the human brain and are capable of reliably classifying even rather noisy input data. Training such a network relies on a large set of training data. As molecular communication as a technology is still in its early development phase, this data is not always readily available. In this paper, we discuss neural network-based demodulation approaches relying on synthetic simulation data based on theoretical channel models as well as works that base their network on actual measurements produced by a prototype test bed. In this work, we give a general overview over the field molecular communication, discuss the challenges in the demodulations process of transmitted signals, and present approaches to these challenges that are based on artificial neural networks.
- ZeitschriftenartikelMachine learning in sensor identification for industrial systems(it - Information Technology: Vol. 65, No. 4-5, 2023) Weber, Lucas; Lenz, RichardThis paper explores the potential and limitations of machine learning for sensor signal identification in complex industrial systems. The objective is a tool to assist engineers in finding the correct inputs to digital twins and simulations from a set of unlabeled sensor signals. A naive end-to-end machine learning approach is usually not applicable to this task, as it would require many comparable industrial systems to learn from. We present a semi-structured approach that uses observations from the manual classification of time series and combines different algorithms to partition the set of signals into smaller groups of signals that share common characteristics. Using a real-world dataset from several power plants, we evaluate our solution for scaling-invariant measurement identification and functional relationship inference using change-point correlations.
- ZeitschriftenartikelFrontmatter(it - Information Technology: Vol. 65, No. 4-5, 2023) Frontmatter
- ZeitschriftenartikelWildfire prediction for California using and comparing Spatio-Temporal Knowledge Graphs(it - Information Technology: Vol. 65, No. 4-5, 2023) Böckling, Martin; Paulheim, Heiko; Detzler, SarahThe frequency of wildfires increases yearly and poses a constant threat to the environment and human beings. Different factors, for example surrounding infrastructure to an area (e.g., campfire sites or power lines) contribute to the occurrence of wildfires. In this paper, we propose using a Spatio-Temporal Knowledge Graph (STKG) based on OpenStreetMap (OSM) data for modeling such infrastructure. Based on that knowledge graph, we use the RDF2vec approach to create embeddings for predicting wildfires, and we align different vector spaces generated at each temporal step by partial rotation. In an experimental study, we determine the effect of the surrounding infrastructure by comparing different data composition strategies, which involve a prediction based on tabular data, a combination of tabular data and embeddings, and solely embeddings. We show that the incorporation of the STKG increases the prediction quality of wildfires.