1 - 10 of 1975
-
PublicationImplementation and evaluation of an additional GPT-4-based reviewer in PRISMA-based medical systematic literature reviews( 2024-09-01)
-
PublicationArtificial Intelligence. What Is Behind the Technology of the Future?(Springer Nature, 2024-05-16)Artificial Intelligence (AI) is already present in our daily routines, and in the future, we will encounter it in almost every aspect of life - from analyzing X-rays for medical diagnosis, driving autonomous cars, maintaining complex machinery, to drafting essays on environmental problems and drawing imaginative pictures. The potentials of AI are enormous, while at the same time many myths, uncertainties and challenges circulate that need to be tackled. The English translation of the book "Künstliche Intelligenz - Was steckt hinter der Technologie der Zukunft?" originally published in German (Springer Vieweg, 2020), this book is addressed to the general public, from interested citizens to corporate executives who want to develop a better and deeper understanding of AI technologies and assess their consequences. Mathematical basics, terminology, and methods are explained in understandable language. Adaptations to different media such as images, text, and speech and the corresponding generative models are introduced. A concluding discussion of opportunities and challenges helps readers evaluate new developments, demystify them, and assess their relevance for the future.
-
PublicationKreativität der generativen KI( 2024-03-28)In diesem Beitrag wird die Frage diskutiert, ob auch Systeme der generativen KI kreative Inhalte erzeugen können. Es wird zunächst beschrieben, wie solche Systeme intern funktionieren und wie sie potenziell neue Inhalte generieren können. Anschließend wird der kreative Prozess diskutiert und es wird überprüft, ob KI-Systeme kreative Leistungen für die unterschiedlichen Medien Text, Bild und Musik erbringen können. In standardisierten Tests konnte gezeigt werden, dass das Sprachmodell GPT-4 inzwischen kreativere Antworten produziert als Menschen. Ähnliche Tests haben ergeben, dass Bilder, die mit einer älteren Version von DALL-E erstellt wurden, nur schwer von Künstlerbildern zu unterscheiden sind. Aufgrund der stark verbesserten Detailgenauigkeit neuerer Systeme ist davon auszugehen, dass diese heute eine verbesserte Kreativität besitzen. Systeme zur Generierung von Musik können derzeit dagegen noch nicht mit der Kreativität menschlicher Komponist*innen mithalten.
-
PublicationIncorporating Query Recommendation for Improving In-Car Conversational Search( 2024-03-23)Retrieval-augmented generation has become an effective mechanism for conversational systems in domain-specific settings. Retrieval of a wrong document due to the lack of context from the user utterance may lead to wrong answer generation. Such an issue may reduce the user engagement and thereby the system reliability. In this paper, we propose a context-guided follow-up question recommendation to internally improve the document retrieval in an iterative approach for developing an in-car conversational system. Specifically, a user utterance is first reformulated, given the context of the conversation to facilitate improved understanding to the retriever. In the cases, where the documents retrieved by the retriever are not relevant enough for answering the user utterance, we employ a large language model (LLM) to generate question recommendation which is then utilized to perform a refined retrieval. An empirical evaluation confirms the effectiveness of our proposed approaches in in-car conversations, achieving 48% and 22% improvement in the retrieval and system generated responses, respectively, against baseline approaches.
-
PublicationUsing ScrutinAI for visual inspection of DNN performance in a medical use case( 2024)
-
PublicationDeep reinforcement learning in service of air traffic controllers to resolve tactical conflicts( 2024)Dense and complex air traffic requires higher levels of automation than those exhibited by tactical conflict detection and resolution (CD&R) tools that air traffic controllers (ATCOs) use today: AI tools can act on their own initiative, increasing the capacity of ATCOs to control higher volumes of traffic. However, given that the air traffic control (ATC) domain is safety critical, requires AI systems to which ATCOs are comfortable to relinquishing control, guaranteeing operational integrity and automation adoption. Two major factors towards this goal are quality of solutions and operational transparency. ResoLver, the system that this article presents, addresses these challenges using an enhanced graph convolutional reinforcement learning method operating in a multiagent setting where each agent – representing a flight – performs a CD&R task, jointly with other agents. We show that ResoLver can provide high-quality solutions with respect to stakeholders interests (air traffic controllers and airspace users), addressing also operational transparency issues, which have been validated by ATCOs in simulated real-world settings.
-
PublicationWasserstein Dropout( 2024)Despite of its importance for safe machine learning, uncertainty quantification for neural networks is far from being solved. State-of-the-art approaches to estimate neural uncertainties are often hybrid, combining parametric models with explicit or implicit (dropout-based) ensembling. We take another pathway and propose a novel approach to uncertainty quantification for regression tasks, Wasserstein dropout, that is purely non-parametric. Technically, it captures aleatoric uncertainty by means of dropout-based sub-network distributions. This is accomplished by a new objective which minimizes the Wasserstein distance between the label distribution and the model distribution. An extensive empirical analysis shows that Wasserstein dropout outperforms state-of-the-art methods, on vanilla test data as well as under distributional shift in terms of producing more accurate and stable uncertainty estimates.
-
PublicationAutomatic scoring of Rhizoctonia crown and root rot affected sugar beet fields from orthorectified UAV images using Machine Learning( 2024)Rhizoctonia crown and root rot (RCRR), caused by Rhizoctonia solani, can cause severe yield and quality losses in sugar beet. The most common strategy to control the disease is the development of resistant varieties. In the breeding process, field experiments with artificial inoculation are carried out to evaluate the performance of genotypes and varieties. The phenotyping process in breeding trials requires constant monitoring and scoring by skilled experts. This work is time demanding and shows bias and heterogeneity according to the experience and capacity of each individual person. Optical sensors and artificial intelligence have demonstrated a great potential to achieve higher accuracy than human raters and the possibility to standardize phenotyping applications. A workflow combining red-green-blue (RGB) and multispectral imagery coupled to an unmanned aerial vehicle (UAV), and machine learning techniques was applied to score diseased plants and plots affected by RCRR. Georeferenced annotation of UAV orthorectified images. With the annotated images, five convolutional neural networks were trained to score individual plants. The training was carried out with different image analysis strategies and data augmentation, respectively. The custom convolutional neural network trained from scratch together with a pre-trained MobileNet showed the best precision in scoring RCRR (0.73 to 0.85). The average per plot of spectral information was used to score plots, and the benefit of adding the information obtained from the score of individual plants was compared. For this purpose, machine learning models were trained together with data management strategies, and the best-performing model was chosen. A combined pipeline of Random Forest and k-Nearest neighbors have shown the best weighted precision (0.67). This research provides a reliable workflow for detecting and scoring RCRR based on aerial imagery. RCRR is often distributed heterogeneously in trial plots, therefore, considering the information from individual plants of the plots showed a significant improvement of UAV based automated monitoring routines.
-
PublicationRechtliche Fairnessanforderungen an KI-Systeme und ihre technische Evaluation( 2024)Im vorliegenden Beitrag werden rechtliche Fairness-Anforderungen an KI-Systeme anhand ausgewählter Kreditscoring-Systeme beleuchtet. Einerseits wird das geltende Recht in den Blick genommen und insbesondere die Frage einer mittelbaren Drittwirkung des Art. 3 GG im Kontext des Kreditscorings erörtert. Andererseits wird auch ein Fokus auf die zukünftige europäische KI-Verordnung (KI-VO) gelegt. Mittels einer juristisch-informatischen Analyse werden Ansätze zur Interpretation und Operationalisierung der fairnessbezogenen Anforderungen an die verwendeten Daten gem. Art. 10 KI-VO nach dem Stand der Technik diskutiert. Diese werden sodann exemplarisch anhand eines konkreten Kreditscoring-Beispiels evaluiert. Der Beitrag schließt mit Empfehlungen hinsichtlich ausgewählter rechtlicher Rahmenbedingungen für einen "fairen" KI-Einsatz.
-
PublicationMaschinelles Lernen einführen: wie MLOps dabei unterstützt( 2023-10-25)Machine Learning Operations (MLOps) bietet eine Orientierung für Entwicklung, Integration und Betrieb von Machine-Learning-Anwendungen. Was MLOps insgesamt auszeichnet.
