Digital Transformation

In the "Smart Inclusion" project, augmented reality is used in the form of smart glasses. It is evaluated to what extent these can be used by people with disabilities to strengthen their individual skills and resources for activities of daily living. The research process will be cooperated and supported by Lebenshilfe Tirol, which is a social institution that offers its residents the opportunity to live and work independently in shared apartments.

Science is clear: Artificial Intelligence will be the defining development of the 21st century. Experts estimate that due to the rise of AI within only 2 decades aspects of daily human life will be unrecognizable. The influence of AI is about to challenge the very organizing principles of our economic and social order. It can generate unprecedented wealth, revolutionize medicine and education, but it can bring existential perils for life as we know it. This makes the EC2020 report that the EU is lagging behind USA and Asia in AI-adoption and development all the more worrying. Among many reasons behind that, the lack of skilled work force is definitively one of the more prominent ones. The objective of AI4VET4AI is to contribute to the digital transformation of the EU labor market by adding new innovative teaching content and methods to VET curricula across 11 European countries and 18 EU NUTS2 regions, in order to support the growth of AI-skilled workers. We start from the ground up: we investigate the most potent sectors for AI-deployment in our 17 regions, and for those sectors, in close cooperation with enterprises and their cluster organizations we create 14 MOOCs and TT materials that can easily be implemented in VET programs (IVET and CVET). We organize 11 innovative AI VET campuses and 7 VET innovation AI incubators, in which VET learners hone their creative and entrepreneurial skills. We use project activities to connect partners closely and to raise awareness of the AI potential in our regions among representatives of public and private sector, as well as civil society. In turn, this helps us to create a joint and active platform of concerned EU citizens and institutions, interested, informed and motivated in supporting AI development further -- this being the basis of our ambitious CoVE - Center of Vocational Excellence, which aims to attract many more institutions (HEIs, VETs, companies, agencies, individuals) in its pursuit of sustainable, inclusive and just AI-powered future for all.

The shortage of skilled workers in the field of software development is damaging competitiveness in Austria as a business location. For this reason, universities are targeting offerings at working professionals, often using blended learning concepts. This involves integrating online and face-to-face teaching in order to offer students increased flexibility. Although blended learning concepts have already proven to be effective, students are confronted with challenges arising from the online component of blended learning approaches on the one hand and the subject matter of programming on the other hand. An Intelligent Tutoring System that provides students with individualized, immediate, and instructive feedback is proposed as a solution.

SMiLe examines the conditions under which solutions using homomorphic encryption are suitable for harnessing the potential of sensitive data for machine learning. The potential of the machine learning techniques is evaluated through two use cases, dealing with employee segmentation and predictive maintenance, respectively. SMiLe contributes to the establishment of a cooperative-creative ecosystem in which various actors interact trustingly, symbiotically, and independently and implement previously unimaginable solutions in which not only data protection and security are guaranteed, but also previously unused potential of data can be exploited. The project will extend existing platforms to provide solutions that enable the creation and use of machine learning models using homomorphically encrypted data.

HLFF Inspiring Minds is a collaboration between the Heidelberg Laureate Forum Foundation and the MCI - The Entrepreneurial School® and is dedicated exclusively to supporting all HLF alumni in developing their personal career paths in mathematics and computer science. Through expert mentoring, coaching and amplified by other activities, bundled on the HLFF Inspiring Minds platform, some of the most aspiring young talents get the unique opportunity to interact and build meaningful relationships with laureates and experts from academia, business and industry.

The Smart Technology Monitoring II project will implement the prototype of an ontology for storing, processing and querying data from partially structured and unstructured data sources. The goal of this project is to test the feasibility of an innovative knowledge management platform that combines information from different technology data sources and must therefore be able to grow and change dynamically. The ontology prototyped in this project is intended to represent an MVP for the core of this software solution, on which future projects can be based.

In the course of an Interreg project "SMART VILLAGES Wipptal" set up for this purpose, a common understanding for a regional portal is to be developed for the communities located in the region (northern South Tyrol with 12 communities and southern South Tyrol with 6 communities). The sub-goal offered by MCI is to survey the logical structure, usability, possible integration options and a specification of the communication platform for the future according to scientific findings and to elaborate it together with the stakeholders.

The project aims to experiment with virtual reality for the training and development of specific competences of students, trainees and apprentices who wish to undertake international mobility or who simply want to try out these experiences virtually. A planned VR platform will host immersive virtual scenarios and other virtual activities, which will allow users to acquire skills to face these new situations that may cause them fear and rejection, weighing negatively on their decision to leave.

CRYSTAL (Conversational Systems for Emotional Support and Customer Assistance) will advance research for a next generation of conversational models and systems that implement reliable, friendly and efficient human-machine interactions that can assist humans in tasks beyond automatic information provision, such as emotional support as well as customer assistance. Emotional support provided to vulnerable individuals, such as ageing adults or people experiencing anxiety or depression. Advanced customer assistance provided to call centres, providing customer care and response to customer inquiries, in an empathic, inviting, accurate and efficient manner. Conversational models and systems will be developed by proposing novel learning strategies for domain adaptation, exploring Large Language Models (LLMs) and prompt engineering methodologies. User state estimation will be also considered, and strategies, such as behavioural models and stakeholder goals, will guide the policies for dialogue management. Novel evaluation frameworks will be also designed by including automatic metrics, usability as well as human acceptance of the technology. Strict ethical and AI robustness compliance procedures will be applied, as conversational recordings will be collected, anonymised, annotated and encrypted for collaborative work among the project partners.

The TensionTerminator (TT) is a certified medical device that helps users relieve tension pain in the back, arm/shoulder belt and neck area effectively and sustainably in a short time by means of self-application. The TT is specially developed for workplace contexts to offer a low-threshold opportunity for employees to relieve their tension pain directly at the workplace. The "smartification" of the TT is intended to allow for an additional level of interaction with users, both on an individual and an organizational level. Thus, the following data will be collected via sensors: (1) force impacts and movement speeds, (2) component function checks, and (3) tracking of the type and frequency of use. Furthermore, in connection with a smartphone app, the patophysiological categorization of the tension situation and its change through the use of the TT will be recorded. The collected data will be processed on the device using a machine learning (ML) based system in order to recognize the type of application, misapplication, sensor errors, etc., to increase the quality of the data obtained, and thus to make subsequent correction loops unnecessary. In real time, this should provide users feedback on optimal and thus effective self-therapy. The aim is to achieve the best possible improvement in the pain and tension situation. Furthermore, the safety of use is to be increased when, for example, indications and warnings are issued in the event of excessive force application, excessive rolling speed, unfavorable force application or improper use. Gamification concepts based on this should encourage employees to use the equipment correctly and prophylactically. Self-motivation is to be increased in connection with notifications, individual leaderboards, linkage with the perceived tension situation, or proof of the improvement in mobility, etc. Furthermore, the sustainable use of the TT provided and its effect on the health and well-being of the employees will be evaluated through ongoing monitoring.

The project "Game Over Eva(sion): Securing Deep Learning with Game Theory" aims to protect deep learning classifiers against targeted attacks. Deep learning classifiers are not only popular in scientific research but are also more and more adapted to the daily life, as self-driving cars, smartphones, and digital personal assistants use these kind of algorithms. Unfortunately, recent research has shown that almost all deep learning classifiers are vulnerable to so-called evasion attacks. In an evasion attack, an attacker can slightly modify a benign object and by this achieves a misclassification with a very high probability. The robustness of deep learning classifiers to these attacks is an open problem. We will model the competition of an attacker who is able to launch an evasion attack and the defender who wants to train a deep learning classifier that is robust against such an attack with means of game theory. In a first step, we will analyze which concepts of other research areas, such as adversarial classification, can be translated to the domain of secure deep learning. Then, we will develop a game-theoretic model that captures all relevant aspects and dependencies between the attacker's and the defender's strategies. In the course of the project, we expect the first theoretically well-founded results on the achievable security of deep learning classifiers in the presence of evasion attacks. Furthermore, we will evaluate existing countermeasures against evasion attacks to gain insights about their optimality when facing a strategic attacker. Finally, we want to implement the key properties from out theoretical models in a practical deep learning classifier. We will compare our classifier against other state-of-the-art classifiers in terms of robustness against evasion attacks and accuracy on benign input objects. By this, we can validate if it is possible to make deep learning classifiers robust against evasion attacks. The expected results of the project will enable us to judge if deep learning classifiers are suitable for scenarios where an attacker has incentives to explicitly fool them, i.e., security-critical areas and widespread consumer products.