Vienna Research Group Leader in the field of "Information and Communication Technology – Interdisciplinary Data Science"

Job Advertisement for Vienna Research Group Leader in the field of "Information and Communication Technology – Interdisciplinary Data Science"

Applications are being invited for outstanding incoming early-career scientists (2-8 years post PhD), interested in building up their independent research group in the field of "Information and Communication Technology – Data Science" at the TU Wien. This group shall strive to improve the understanding of substantial current scientific research questions in the field of data science that lead to applicable solutions in and together with other disciplines with potential medium term economic or social benefits. The ability to bridge disciplines is thus a key evaluation criterion. New interdisciplinary combinations are also encouraged and highly appreciated.

The aim of this announcement is to look for and attract exceptional candidates, who, once selected, will then go on to submit an application in tandem together with an experienced scientist at TU Wien, for the call “Vienna Research Groups for Young Investigators” by the Vienna Science and Technology Fund (WWTF): https://www.wwtf.at/programmes/vienna_research_groups/#VRG19

In the case of a successful funding decision by the WWTF bord, the research group will be financed for 6 – 8 years, with up to 1.6 million EUR being provided by the WWTF, and an additional contribution from TU Wien itself. The successful candidate will be offered a tenure-track position at the TU Wien.

Requirements:

Applicants coming from abroad should have exceptional promise, or a proven outstanding record of research achievement, within the field of “ICT - Interdisciplinary Data Science”. They should also provide strong evidence of their potential to make a significant contribution to substantial state-of-the-art scientific research questions in this particular research field. Female applicants are explicitly encouraged to apply.

Application procedure:

Applications have to include a CV with publication list, list of relevant research projects, teaching activities (incl. evaluations, if available) and supervised PhD students as well as a short statement of the intended research project in form of a motivation letter. Please send all documents to wwtf@tuwien.ac.at with the subject “Research Group Title/ TU Wien host” to allow a clear assignment. Application deadline is May, 2nd 2019 latest. All applications will be treated confidentially and according to the Data Protection Guideline.

For further information please contact:

  • Scientific questions: the respective group hosts – please see “scientific contact” in each group description
  • Administrative issues: Elisabeth Schludermann, Funding Support and Industry Relations, Research and Transfer Support: elisabeth.schludermann@tuwien.ac.at

Possible hosting labs at TU Wien

(in alphabetical order)

Scientific contact person: Clemens Schwaiger, Email
Webpage:  http://www.ift.at

The Institute of Production Engineering and Photonic Technologies (IFT) is closely connected with leading production companies in Austria and beyond and creates innovations in cooperative research in the field of machining technology. Due to this extensive network of research institutions and frontrunner companies, research projects at IFT aim at improving existing manufacturing processes and developing new cutting-edge technology solutions.

Large-scale data acquisition is on the rise due to falling storage cost and the cheapening of high quality sensors. This fuels the increasing complexity of production machines parallel to soaring interconnectivity, which manifests itself in combination in the form of so-called cyber-physical systems. In order to make use of the large amounts of data, popularly denoted under the term big data, manually performed analyses cannot deliver satisfactory results due to high volume, velocity and variety of the captured datasets. It is necessary to implement AI-based solutions in the manufacturing domain to not only handle complex systems, but also to gain additional information, which in the first step will be used to provide assistance in human decision making by providing prognostics and visualization (human-in-the-loop).

The new research group will be on par with the existing groups Manufacturing Technology, Machine Tools, Production Measurement Technology and Adaptronic Systems and Control Technology and Integrated Systems. It will initially build on ongoing projects and consolidate all relevant works inside the IFT in an effort to facilitate the implementation of artificial intelligence in the manufacturing domain on a broad basis. Austria is not a newcomer in the area of AI, considering the development of the Long Short-Term Memory (LSTM) architecture by Sepp Hochreiter in 1997 or recent developments like the Equation Learner (EQL) by Martius and Lampert in 2016. Many possible applications for AI in the manufacturing domain have already been investigated, especially in the AI-subdivision of machine learning. In order to remain at the cutting edge of technology development, current advances on deep learning and efforts to overcome the black box characteristic of machine learning will be embedded in applications inside the manufacturing domain. The new research group will therefore assume an interface function between production technology and data science.

 

Scientific contact person: Geraldine Fitzpatrick, Email 
Webpage: http://igw.tuwien.ac.at/hci/

Recent research and mainstream media reports point again and again to the unexpected consequences, unintentional biases and complex challenges of using 'big data'. Bad data science has significant societal, ethical and human consequences. As Moshe Vardi recently stated, "Suddenly we [computer scientists] are running society and we are poorly equipped." As Human-Computer Interaction researchers, we have the interdisciplinary expertise to respond to these challenges in the holistic ways necessary. We work at the intersection of designing technology and understanding societal visions and desirable futures. We often act as catalyst between what is technically possible with data science and what is ethically acceptable, bringing human and societal values into design processes. This expertise is honed from decades of responding to the human-social scale challenges of each new wave of technology innovation since the 1950s. We have the conceptual and methodological tools and theoretical foundations to understand people and contexts in their relationship with new technological opportunity spaces. We believe that bridging the gap between innovation and societal responsibility is precisely what is needed in response to the new challenges entailed in this data science wave. It also presents a unique opportunity for the European Innovation and Research arena to position itself with a unique perspective on the global competition for the best ideas.

Scientific contact person: Robert Sablatnig, Email
Webpage: http://igw.tuwien.ac.at/hci/

Recent research and mainstream media reports point again and again to the unexpected consequences, unintentional biases and complex challenges of using 'big data'. Bad data science has significant societal, ethical and human consequences. As Moshe Vardi recently stated, "Suddenly we [computer scientists] are running society and we are poorly equipped." As Human-Computer Interaction researchers, we have the interdisciplinary expertise to respond to these challenges in the holistic ways necessary. We work at the intersection of designing technology and understanding societal visions and desirable futures. We often act as catalyst between what is technically possible with data science and what is ethically acceptable, bringing human and societal values into design processes. This expertise is honed from decades of responding to the human-social scale challenges of each new wave of technology innovation since the 1950s. We have the conceptual and methodological tools and theoretical foundations to understand people and contexts in their relationship with new technological opportunity spaces. We believe that bridging the gap between innovation and societal responsibility is precisely what is needed in response to the new challenges entailed in this data science wave. It also presents a unique opportunity for the European Innovation and Research arena to position itself with a unique perspective on the global competition for the best ideas.

Scientific contact person: Hannes Werthner, Email 
Webpage http://ec.tuwien.ac.at/

The e-commerce Research Unit deals with Data Science / data analytical methods in e-commerce and media, more specifically we explore and want to understand consumer and reader behavior using methods such as text and sentiment mining, network analysis, machine learning. Research is both applied as well as fundamental.

Scientific contact person: Peter Filzmoser, Email
Webpage: http://cstat.tuwien.ac.at/filz/

The group "Computational Statistics" has its focus on developing new statistical Methods and algorithms for handling non-standard, large and complex data. The group is specialized on developing methodology for data with outliers or noisy data, imbalanced data, data containing many non-informative variables, and non-standard data such as compositional data. The new group will complement the "Computational Statistics" Group particularly with the computing aspect as well as with applications in Industry 4.0,  Robotics and Biomedicine. Their background will be computer science, with strong Focus on statistics and machine learning. Jointly, we will develop methods and algorithms for near-to-real-time modeling, which is important in many manufacturing processes, but also in biomedical applications (e.g. intensive care medicine). New approaches in the spirit of extreme learning machines will be developed to reduce the long training phase of methods such as deep learning. Algorithms for robust statistical methods will be extended to be applicable in online situations.

Scientific contact person: Nysret Musliu, Email
Webpage: https://www.dbai.tuwien.ac.at/staff/musliu/

Our research is focused on developing problem solving techniques to solving various complex real-world problems. These techniques are based on artificial intelligence methods such as heuristic search, machine learning, constraint programming, SAT and hybrid approaches. Considerable attention is being paid to the automation of problem-solving techniques with regard to algorithm selection/configuration and hyper-heuristics which automate the design of heuristic methods based on machine learning. The application areas include planning and scheduling, educational timetabling, logistics and other business fields.

Scientific Contact Person: Franz Hlawatsch, Email 
Webpage: http://www.nt.tuwien.ac.at/

In metropolitan cities, the huge size of data collected by handheld devices, vehicles, and the Internet of Things creates new opportunities but poses challenges to efficient data processing, storage, and dissemination. The goal of the envisaged research group is to leverage big data by establishing a new data science framework for urban situational awareness. Principles and methodologies from inference, learning, and control will be synthesized with a focus on big data processing. The resulting data science framework will accelerate the development of emerging services and applications including indoor localization and mapping, universal transportation, autonomous navigation, and public safety. This will help establish Vienna as a key venue for data science in an urban setting.
The research group will be embedded at the Institute of Telecommunications, TU Wien.

Scientific contact person: Allan Hanbury, Email
Webpage: http://ec.tuwien.ac.at/DataIntelligenceLab

Data Science Methods are generic - they can be applied to produce results with any data, even if the results do not make sense in the real world. When using Data Science methods to support other disciplines, such as Medicine, Law, and Innovation, it is currently a challenge to effectively elicit and incorporate existing domain knowledge to guide the Data Science approaches and make them transparent, while at the same time not being put off course by biases that commonly exist in domains in various forms, such as within the data or accepted practice.

Scientific contact person: Klaus Nordhausen, Email
Webpage http://klausnordhausen.com/


The group "Computational Statistics" has its focus on developing new statistical methods and algorithms for handling non-standard, large and complex data. The group is specialized on developing methodology for data with outliers or noisy data, imbalanced data, data containing many non-informative variables, and non-standard data such as compositional data and tensorial data. Also dimension reduction and blind source separation is an important area of research of the group. The new group will complement the "Computational Statistics" group by adding Gaussian processes to the research portfolio of the group. Gaussian processes are interesting from a theoretical and applied point of view and the new group will derive theoretical properties of such processes, consider their covariance function estimation and develop new kernels. A special focus will be given to large data sets, functional data and sequential procedures as well as deep Gaussian processes. Focus will be here also on the robustness properties of these processes. Gaussian processes are well established in machine learning, the design and analysis of computer experiments and engineering applications and therefore of huge interest for a technical university with many possibilities for interdisciplinary collaborations and Research. 

Scientific contact person: Heinz Pettermann, Email
Webpage: https://www.ilsb.tuwien.ac.at/ilsb/lightweight-design/

Modern high performance materials, both technical and biological ones, show highly nonlinear and direction dependent behavior. Investigation of their response is done by 'micromechanics of materials' approaches with great success. However, such methods are less suitable for the analysis of large structures. For that, constitutive material laws are desired based on numerical efficient phenomenological mathematical models. Since conventional approaches suffer from simplifications or lead to complicated formulations, there is high potential for machine learning methods (and the like) in this field.
The Institute of Lightweight Design and Structural Biomechanics has great experience in materials modeling and simulation and can provide detailed, highly resolved predictions of the response of e.g. composite materials. There are computational tools for the generation of models, their simulations, and proper post-processing to extract required training and verification data in high number in an automated way. Such predictions are used as input for machine learning algorithms with the goal to deduce the constitutive behavior for those materials. The trained artificial intelligence framework is expected to be very efficient in terms of numerical effort to gain predictions and can be ideally combined with structural analyses software, e.g. Finite Element Methods. This will enable simulations of the nonlinear response of large structures, which are out of scope with common state of the art methods.
In addition, this approach can be applied to problems with uncertainties in input data on the field of mechanics of materials and structures.

Scientific contact person: Marko D. Mihovilovic, Email
Webpage http://www.ias.tuwien.ac.at/home/

Researchers in my department, the Institute for Applied Synthetic Chemistry, are experts in designing and creating molecular structures that enable new functions of technological or biological relevance. We develop materials for dental fillings, create new biocatalysts and produce bulk chemicals, perform organic reactions in living cells, and develop compounds as pharmaceutical drug candidates, among many other areas.

Scientific contact person: Andreas Rauber, Email
Webpage: http://www.ifs.tuwien.ac.at

The Research Unit on Information and Software Engineering (ifs) performs both foundational and applied research as well as development at the confluence of information and software, with a focus on providing means to allow us to trust the complex socio-technical systems we create and interact with. This includes, specifically, research in data analytics driven approaches for structured and unstructured data, in particular, text and media information. Research into novel machine learning and AI techniques as well as modelling and simulation based approaches aim at improving both the performance of the data analytics capabilities as well as at increasing their transparency and explainability to ensure we can trust the insights and decisions stemming from complex data science processes. With a focus on reproducibility as a key requirement for solid R&D, IFS performs research on the establishment of solid research infrastructures to support the research work in this unit and beyond, establishing technical and organizational solutions to data stewardship and research data management. We investigate mechanisms that ensure that solutions are sustainable and scale up to the requirements of Big Data analytics and meet the requirements of privacy-preserving data analysis and management of highly sensitive data, integrating activities from the aforementioned topic areas. The new research group will strengthen and expand the research activities of the Research Unit IFS in the areas listed above, with a specific focus on interdisciplinary research activities. The candidate is expected to have an excellent research record in one specific or at the intersection of the core fields of data science, i.e. machine learning, algorithmic accountability and transparency, research data management, semantic technologies, or big data infrastructures. The candidate further should demonstrate experience in cooperative research with other disciplines (in any domain or across different domains, ranging from astronomy, via chemistry, climate data, finance, genetics, with a preference towards research domains established at TU Wien or at other research institutions established in Vienna).

Scientific contact person: Silvia Miksch, Email 
Webpage: https://www.cvast.tuwien.ac.at/

Visual Data Science incorporates Visual Analystics and Visualization to guide, steer, and support the analytical reasoning process in data science. Visual Analytics is “the science of analytical reasoning facilitated by interactive visual interfaces” (Thomas & Cook 2005) and aims to make complex information structures more comprehensible, facilitate new insights, and enable knowledge discovery. Visual Analystics leverages the specific strengths of computers and humans for the best possible outcome: on the one hand, computers are better at managing and processing large amounts of data by exploiting their enormous computational power; on the other hand, humans have better perceptual and cognitive means, which enable them to visually perceive unexpected patterns and to interpret data. It is a multidisciplinary approach, integrating aspects of statistics, data mining and knowledge discovery, information visualization, human-computer interaction, and perceptual and cognitive science.