Krause, A. F. *, Kannen, K. *, Büscher, S., Ressel, C. & Wild-Wall, N. (2025). Pro-adaptive Cognitive Assistive Technology: Concept and Application in Reading Support for ADHD. In International Conference on Extended Reality (pp. xx-xx). Springer Lecture Notes in Computer Science. (in press).
(*) These authors contributed equally to this work.
Buschmeier, H., Hassan, T., & Kopp, S. (2024, November). Multimodal Co-Construction of Explanations with XAI Workshop. In Proceedings of the 26th International Conference on Multimodal Interaction (pp. 698-699).
The ICMI 2024 workshop on “Multimodal Co-Construction of Explanations with XAI” bridges the fields of Explainable Artificial Intelligence (XAI) and Multimodal Interaction, focusing on the recent perspective that effective AI explanations should be dynamically co-constructed through interactive, social processes involving both the explainer and the explainee. By framing XAI explanations as a multimodal, interactive co-construction challenge, the workshop seeks to explore how these two fields can collaboratively address the complexities of creating understandable and context-sensitive XAI systems.
Yavuz, S. (2024). Development of a 2-4 double arbiter PUF design on FPGA with enhanced performance.
Implementation of delay-based Physical Unclonable Functions (PUFs) on FPGAs poses significant challenges due to high requirements, such as the generation of unique and reliable keys. These requirements must be fulfilled, especially when using PUFs in security applications, otherwise security cannot be guaranteed. In addition, it must be ensured that physical disturbances such as fluctuations in the ambient temperature do not have a major impact on the performance of the PUF and therefore on security. In this paper, the implementation and evaluation of a novel 56-bit 2-4 Double Arbiter PUF (DAPUF) is presented. For performance analysis, the proposed 2-4 DAPUF is investigated with 20 Digilent Nexys-3 (AMD-Xilinx Spartan-6 FPGA) boards and a large data set of 30 million challenges under varying ambient temperature in the range from 0°C to 50°C. Our experimental results show that the proposed 2-4 DAPUF is resistant to temperature fluctuations. Here, the maximum change in reliability amounts to 1.18%. For randomness and uniqueness, the changes are less than 0.50%. Furthermore, our results show that performance can be significantly improved by combining PUFs with XOR operations.
Yavuz, S., Daniel, K., & Naroska, E. (2024). Vulnerabilities and challenges in the development of PUF-based authentication protocols on FPGAs: A brief review.
The security of IoT (Internet of Things) devices and the protection of sensitive information processed by these devices such as personal data, sensor values, process-related information is an important and difficult challenge. A major task in IoT communication is secure identification of devices. Unfortunately, traditional cryptographic methods are often not suitable for IoT devices due to their limited hardware resources. However, typical methods are computationally intensive, require a large amount of memory, and have a high-power consumption. On the other hand, Physical Unclonable Functions (PUFs) are low-cost and lightweight hardware-based primitives that can be used as a security component to protect data against third parties and thus increase security of a device. In this paper, the challenges and vulnerabilities in the development of PUF-based authentication protocols are presented. To this end, a security analysis of different approaches known from literature are discussed. Furthermore, possible attack vectors and prevention techniques are also considered.
Stolarz, M., Romeo, M., Mitrevski, A., & Plöger, P. G. (2024, August). Deep Learning-Based Adaptation of Robot Behaviour for Assistive Robotics. In 2024 33rd IEEE International Conference on Robot and Human Interactive Communication (ROMAN) (pp. 110-117). IEEE.
Robot behaviour models in socially assistive robotics are typically trained using high-level features, such as a user’s engagement, such that inaccuracies in the feature extraction can have a significant effect on a robot’s subsequent performance. In this paper, we study whether a behaviour model can be meaningfully represented using an end-to-end approach, where multimodal input, concretely visual data and activity information, is directly processed by a neural network. This paper concretely analyses the different building blocks of such a model, such that the aim is to identify a suitable architecture that can meaningfully combine the different modalities for guiding a robot’s behaviour. We conduct the analysis in the context of a sequence learning game, such that we compare different vision-only models that are then combined with an activity processing network into a joint multimodal model. The results of our evaluation on a dedicated dataset from the sequence learning game demonstrate that a multimodal end-to-end behaviour model has potential for assistive robotics — we report an F1 score of around 0.88 across different dataset-based test scenarios — but the real-life transferability strongly depends on whether the data is diverse enough for capturing meaningful variations in real-world scenarios, such as users being at different distances from a robot.
Gjoreski, M., Hassan, T., Vered, M., Houben, S., & Kopp, S. (2024, October). XAI for U: Explainable AI for Ubiquitous, Pervasive and Wearable Computing. In Companion of the 2024 on ACM International Joint Conference on Pervasive and Ubiquitous Computing (pp. 992-995).
The workshop XAI for U aims to address the critical need for transparency in Artificial Intelligence (AI) systems that integrate into our daily lives through mobile systems, wearables, and smart environments. Despite advances in AI, many of these systems remain opaque, making it difficult for users, developers, and stakeholders to verify their reliability and correctness. This workshop addresses the pressing need for enabling Explainable AI (XAI) tools within Ubiquitous and Wearable Computing and highlights the unique challenges that come with it, such as XAI that deals with time-series and multimodal data, XAI that explains interconnected machine learning (ML) components, and XAI that provides user-centered explanations. The workshop aims to foster collaboration among researchers in related domains, share recent advancements, address open challenges, and propose future research directions to improve the applicability and development of XAI in Ubiquitous Pervasive and Wearable Computing - and with that seeks to enhance user trust, understanding, interaction, and adoption, ensuring that AI- driven solutions are not only more explainable but also more aligned with ethical standards and user expectations.
Schneider, J., Cheruvalath, S. S., & Hassan, T. (2024, October). Time for an Explanation: A Mini-Review of Explainable Physio-Behavioural Time-Series Classification. In Companion of the 2024 on ACM International Joint Conference on Pervasive and Ubiquitous Computing (pp. 885-889).
Time-series classification is seeing growing importance as device proliferation has lead to the collection of an abundance of sensor data. Although black-box models, whose internal workings are difficult to understand, are a common choice for this task, their use in safety-critical domains has raised calls for greater transparency. In response, researchers have begun employing explainable artificial intelligence together with physio-behavioural signals in the context of real-world problems. Hence, this paper examines the current literature in this area and contributes principles for future research to overcome the limitations of the reviewed works.
Krause, A. F., Essig, K., Wild-Wall, N., Ressel, C. (2024), A proposal for the concept of Pro-adaptive Cognitive Assistive Technology. Abstract, accepted at Human Factors and Ergonomics Societey Europe (HFES).
Assistive Technology is becoming an integral part of our daily live, supporting people in different areas, for example while driving a car or cognitive demanding tasks at work or home. Yet, existing Assistive Technology often only considers the current situational context and capabilities of a user. Here, we propose the concept of “Pro-adaptive Cognitive Assistive Technology” (Pro-CAT), that adapts to predictable, temporal changes in the users capabilities and contextual situation (e.g., general ageing processes, an expected course of a disease, learning progress during skill aquisition or environmental changes). Pro-CAT can have several advantages:
Wild-Wall, N., Ressel, C., Kannen, K., Krause, A. F., Büscher, S., Mosler, B., & Arntz, B. (2024). Strukturen zur Berücksichtigung ethischer Aspekte in der Entwicklung von digitalen assistiven Technologien für vulnerable Gruppen. In K. Hegemann, D. Lud & F. Sohnrey (Hrsg.), Wer rettet die Welt? Transformation gestalten in Zeiten der Polykrise (S. 165–182). Nomos.
Ferger, A., Krause, A. F., & Pitsch, K. Workflows and Methods for Creating Structured Corpora of Multimodal Interaction. 14–15 September 2023, University of Mannheim, Germany, 73.
Corpus analysis of computer mediated and/or multimodal interaction can draw on methods of written and spoken corpora, while also providing further information like gaze or walk annotations or sensor-based data like kinect or motion capture or robot log files. We propose a workflow leveraging the developments of both worlds while simultaneously focussing on standard formats and a sustainable way of research data management.
