Objectives
In the recent past, the academic discipline of computer science has seen very rapid advancements, and especially its impact and options for use have expanded dramatically. The topics and methods of computer science have meanwhile influenced virtually all other academic disciplines, and in addition, its applications have manifold effects on society and all spheres of life. This development of computer science – from an academic discipline originally characterised by its orientation towards technological engineering to a discipline covering a very wide range of methodological approaches, as well as numerous applications and effects – has opened up many new links and connections particularly in the context of the broad spectrum of subjects of the University of Vienna, which also encompass, for instance, the social sciences, humanities and human science. The (potential) impact of ‘artificial intelligence’ (AI) has a strong influence on virtually all areas of application of digital technologies. Computer science plays a leading role in algorithmic methodology and in the technologies on which AI is based.
The Faculty of Computer Science at the University of Vienna has, in recent years, specifically focused on identifying and utilising the potential connected with these developments. In particular, the Faculty has endeavoured to develop a topical profile, visibility and critical mass in four areas, i.e. in the key research areas of algorithms, data science, systems and human-centred computing. In order to also further the necessary close links between the rapid methodological and conceptual progress in the area of AI on the one hand and the application of AI in all academic areas of the University of Vienna on the other, the Faculty of Computer Science regards it as its responsibility to contribute its expertise as a point of call and as a partner for exchange and cooperation in all questions arising with regard to AI regarding research, teaching and the Third Mission at the University of Vienna. During the next development stage, it intends to continue pursuing this successful path.
There needs to be a balance between basic research and applied research. Technology transfer activities contribute to the sustainable impact of research activities.
Strength in the discipline (and the resulting visibility) are key prerequisites for a further objective of the Faculty of Computer Science: building and advancing interdisciplinary cooperation in the context of the University of Vienna. By means of well-defined forms of cooperation with areas of application of digital technologies, the crucial components of computer science that are oriented towards basic research also provide essential input for tackling major social challenges of our time. Based on the existing strong points of the Faculty of Computer Science, in the next development stage the focus at the Faculty will be especially on the potential for enhancing interdisciplinary cooperation and research activities in three areas:
- with the biosciences and life sciences (based on expertise in the key research areas of algorithms and data science, for instance in the fields of bioinformatics, neuroinformatics and computational drug design);
- with business, economics and statistics (based on expertise in the key research areas of systems and data science, for instance in the areas of business informatics and security);
- with the social sciences and the humanities (based on expertise in the key research areas of human-centred computing and data science).
Key Research Areas
Algorithms
The key research area of algorithms focuses on the development, analysis and improvement of algorithms. It studies various types of questions, particularly common algorithmic questions regarding networks, which can often be modelled as graphs, as well as the algorithmic basis of machine learning and thus of AI. Questions regarding networks are of key significance in numerous areas of application, for instance in social networks, communication networks, which constitute the backbone of our digital society, or in cloud data centres and super computers, which can meanwhile comprise millions of processors within a system. In all areas, new algorithms are needed which meet high efficiency and scalability requirements: For instance, due to the popularity of data-centred applications (in the areas of health, business, social networking, etc., as well as, of course, AI), data traffic has seen an enormous increase, so that many networks will soon reach their capacity limits. Efficient use of energy and sustainability are also controlled at the level of algorithms and thus represent an additional important aspect of research.
Many issues concerning large networks require solutions to algorithmic problems on the basis of graphs. In this key research area, efficient graph algorithms are being developed, analysed theoretically and also assessed empirically. The research activities in this area also include dynamic, distributed and parallel as well as numerical algorithms. Further research activities closely related to this field, with important links to the key research area of data science at the Faculty, focus on questions concerning algorithmic components of AI (e.g. neural networks), as well as algorithms for the generation of knowledge from social networks. Graph-based abstractions also serve as the basis for algorithms and for programming future computer architectures, which are massively parallel on the one hand, but, for reasons of energy efficiency, also increasingly heterogeneous on the other. In task-based runtime systems, complex scalable and adaptive algorithms that are used as a basis for computationally intensive and data-intensive applications can be represented as dynamic graphs. These technologies play a key role for developing a new generation of parallel programming models.
Data Science
Data science focuses on acquiring knowledge from data. Due to the digital transformation, methods of data science have meanwhile become necessary in almost all academic disciplines. In addition, data-driven research plays a key role in numerous academic fields: For instance, in the biosciences and life sciences or in pharmacy, as well as in the humanities and the social sciences, new insights have increasingly often been based on methods of data science. At the same time, questions from other academic areas have inspired the development of new data science methods. In its broadest interpretation, data science is an interdisciplinary research field with an inherent bridging function because an intensive cooperation between those developing new methods and those applying them is required. A wide range of interdisciplinary research activities in this area are being conducted at the University of Vienna. The Faculty of Computer Science pursues research regarding central methodological components of data science, and plays a leading role in the relevant interdisciplinary activities, where it contributes its expertise in computer science.
Since datasets are growing continuously and very rapidly, the use of computer science methods to generate knowledge from data is an indispensable cornerstone of data science research. The research topics in this field encompass the entire process of knowledge generation from data: database research techniques for efficient storage, representation, organisation and similarity search of very large volumes of data; data mining methods for detecting trends and patterns; machine learning methods and AI methods for forecasting correlations (interpretable machine learning and robust machine learning are of particular interest in this regard); visualisation methods for understanding data and models. Here, links with the algorithmic methodological components of computational science exist, where traditional ab-initio models are being increasingly supplemented by data-driven models, so that the use of machine learning methods has also become very important.
Data science is an emerging research area, since increasing amounts of data can be acquired and made available digitally in almost all areas of knowledge, and computing infrastructure has seen a rapid advancement over the past few decades. However, the constant further development and diversity of the computing infrastructure also requires a permanent advancement of algorithms and runtime systems, as well as of tools and software in order to meet the ambitious goals that have been set in data science. Close links with activities at the key research area of algorithms at the Faculty have therefore been established – for instance, with regard to robustness and scalability of numerical algorithms, methods for analysing neuronal data, text mining, and software or middleware.
Systems
This key research area of the Faculty is based on the observation that the real and the digital worlds will converge further in the future. In this context, the focus is primarily on systems that are needed in this process of transformation. The challenge here is to explore and develop methods and processes concerning knowledge in intelligent systems, as well as knowledge on intelligent systems, taking new approaches into account.
This results in research questions such as the following: How can systems be developed and modelled in line with a design-oriented approach in order to enable new architectures in a disruptive environment (sustainability)? How can domain-specific knowledge be formalised, and how can a representation thus become machine-understandable (operationalisable, intelligent)? How can the behaviour of these intelligent systems be designed in a comprehensible way (explainability)? How can security and privacy be ensured in this context (secure systems)? How can we meet the challenge of the constantly increasing distribution of information systems (distributed systems)?
The complexity and diversity of the topics arising due to digitalisation in general and AI in particular is addressed, on the one hand, by establishing the corresponding research approaches and, on the other, by design-oriented approaches that also take disruptive technologies into account. The above research questions are of key relevance to the core of computer science, and they are an essential component of modern systems-oriented and design-oriented business informatics at the Faculty of Computer Science.
This key research area of the Faculty comprises research on, and the development of, approaches, methods and tools for areas such as cloud computing, flexible and distributed processes, parallel computing, conceptual modelling, intelligent and agile agents, DevOps, semantic technologies, the Internet of the future, secure and service-oriented systems, cooperative systems, IT infrastructure for Industry 4.0, cyber-physical systems (CPS), the Internet of Things and information security management systems, as well as blockchain systems.
Human-Centred Computing
The key research area of human-centred computing puts special emphasis on human beings and their diverse needs and aspirations. It encompasses both the theoretical and the experimental development of human-computer systems, interfaces, models and interactive processes. Human-centred computing is an inherently interdisciplinary area with an impact on many other disciplines, particularly on the social sciences and on the humanities, and, in the context of digital humanism, it is also concerned with political, ethical and aesthetic questions of digitalisation. In view of the rapid technological progress in the area of AI in the present day, this research area is significantly gaining in relevance as a particularly important complement to the research activities in the other three key research areas at the Faculty. The links to (school) education also play a special role due to the impact of AI on the fields of education and didactics.
The corresponding key research area at the Faculty of Computer Science focuses on the computer science and informatics aspects of human-centred computing and the resulting links to other disciplines. At the centre of research activities lies the vision that human-centred computing can contribute to improving quality of life, social inclusion, effectiveness, as well as personal fulfilment and purpose at the individual and the social levels, and to reducing the digital gap. Members of the Faculty study the human-centred design of human-computer interfaces, assistive communication devices, brain-computer interfaces, technologies and systems for the inclusion and empowerment of people with impairments, the improvement of user experience to increase the acceptance of applications, and questions of the values-based use and sustainable advancement of information and communications technology.
In addition to the focus on designing interfaces between human beings and computers, the Faculty of Computer Science is engaged in several initiatives, including interdisciplinary initiatives, to examine technologies that aim to assist human beings in regard to learning, decision-making, working and improving their quality of life in the context of digital change. This takes place, for instance, by means of comprehensible explanations and visualisations of AI models, empirical studies of factors influencing the use of information technologies, by expanding and improving human learning processes through digitally supported access, as well as by means of digital technologies supporting communication and cooperation.
Important links have been established between the key research areas of human-centred computing and data science with regard to technology-enhanced learning and data visualisations that are easy to grasp for human beings. The key research areas of systems and of human-centred computing are inherently linked through the research approach of design thinking, aimed at developing human-centred systems and interfaces. In addition, security and the protection of privacy are further essential characteristics of human-centred computing, and thus closely linked with the security concerns of the key research area of systems regarding usable security. Further important links have been established between the key research area of human-centred computing and the research area of computer science education (didactics of computer science), which focuses on the discipline of computer science in educational contexts, particularly by means of physical computing, gaming education and technology-enhanced learning.