Research

Our research focuses on the intersection of Visual Analytics, Information Visualization, HCI, Interactive Machine Learning, and Explainable AI with a human-centered approach to data-driven decision-making. We address complex challenges in data, model, and human interaction, ensuring that humans are in the loop when needed. Our interactive visual data analysis and human-AI collaboration solutions typically enhance the “E”s: data Exploration, model Explanation, and human knowledge Externalization, improving decision-making in application domains like digital health, digital humanities, digital libraries, finance, and sustainability research.

The Dynamic and Distributed Information Systems Group investigates how to combine human and artificial intelligence (AI) to address profound societal challenges. It investigates the engineering foundations of artificial intelligence that rely on multi-modal knowledge, the design/implimentation of collectively intelligent systems, and the evaluation of such systems in real-world settings such as digital democracy, the future of new media, or medicine.

The HASE Lab focuses on the human aspects of software engineering to boost developer productivity and well-being. The research includes: (1) empirically studying developers' perceptions and factors of productivity and well-being, (2) the use of physiological and computer interaction sensors to measure cognitive and emotional states, and (3) developing approaches to improve the developer experience and flow in the workplace.

The Software Evolution and Architecture Lab (s.e.a.l.) at UZH focuses on analysis of software systems, with a specific emphasis on their long-term evolution, defect prediction, and quality analysis. The expertise covers software engineering with AI, data mining large repositories about software development, software architecture and design evolution, software quality assessment, and empirical studies to help increase productivity of software developers. The group collaborates with the world of practice and has strong ties with multi-national software corporations.

The AIML group explores machine learning techniques, especially deep learning for image processing in general, face recognition in particular and other image classification tasks.
We focus on comparability of algorithms, robustness against unknown data, stability against adversarial attacks, fairness of algorithms with respect to minority groups, explainability of predictions, and the incorporation of well-known image processing and machine learning techniques into deep learning frameworks.
We collaborate with many partners throughout the Zurich universities, other Swiss research groups, and internationally, on topics like medical data processing, activity recognition from smartwatches, and neuroinformatics.
The group is also tightly coupled with the Digital Society Initiative and the School for Interdisciplinary Studies of the University of Zurich, as well as the Center for Reproducible Research.
All of our research is conducted as open source and reproducible research, and we publish in top conferences and journals in computer vision and machine learning.

We are an interdisciplinary research group studying the impact of digitization on people and society, particularly interested in problems that arise in automated decision-making. In the fast-paced online environment, companies observe user behavior and utilize big data algorithms to personalize content, perpetuating social biases as these algorithms learn from human data. Our projects focus on both uncovering and measuring the problems imposed by these large algorithmic systems, as well as coming up with mitigation strategies and policy recommendations.

We conduct Human-Computer Interaction research with a focus on designing and understanding the impact of technology for social good. Our main threads of exploration center around the role of computing in fostering mental and physical health,  environmental sustainability, education, and diversity and inclusion in regard to gender, culture and disability. Our methods combine computer science with approaches drawn from cultural anthropology, sociology, psychology, and design.

Our mission is to understand computational challenges for data processing and design simple and scalable solutions towards these challenges. Theoretical research includes the development of novel data processing algorithms and languages along with the analysis of their complexities. Systems research is on building data systems in academia and industry based on well-understood theory. Sample of research themes: in-database linear algebra and machine learning; adaptive static and dynamic query processing; factorized databases; and probabilistic databases.

The reconstruction of as-built interiors of buildings is a challenging research problem. An effective approach must be able to faithfully capture the architectural structures and separate permanent components from clutter (e.g. furniture), while at the same time dealing with defects in the input data. To solve this ill-posed problem, we develop efficient approaches to reconstruct 3D interiors.

The Autonomous Learning and Predictive Intelligence (ALPI) group works to build new algorithms to solve sequential decision-making problems. The research interest mostly focuses on theoretical Reinforcement Learning, Imitation Learning, and Multi-Agent learning, but we are interested also in every principled learning algorithm. With a commitment to creating reliable and efficient solutions, our ultimate goal is to address real-world challenges. Our works are presented at major machine-learning conferences such as NeurIPS, ICML, ICLR, AAMAS, AAAI, or AISTATS.

The Robotics and Perception Group led by Prof. Davide Scaramuzza researches robust deep learning, perception, and control methods to enable autonomous navigation of agile robots, such as drones, in challenging real world environments using only onboard perception and computation.

The Information Management Research Group focuses on fruitful collaboration, applying the newest digital technologies for business or government purposes. We study and improve advice giving (e.g. in the financial sector), small and large group collaboration and inter-organizational collaboration. In the past we developed and applied tools based on advanced interaction technologies or Blockchains. Currently, we engage in developing digital agents. This includes the application of generative AI and drones.

The Computation and Economics Research Group does research at the intersection of computer science and game theory, with a particular focus on market design. We are interested in questions such as how to design market institutions that work well, or how to repair market platforms that are broken. In the analysis and design of marketplaces, we employ mathematical analysis, computational/algorithmic techniques, lab/field experiments, and simulation techniques, taking into account the incentives of market participants modeled using microeconomic theory and game theory. We are interested in a wide variety of market domains, including financial markets, spectrum auctions, electricity markets, cloud computing markets, school choice systems, and online trading platforms.

The Communication Systems Group CSG established excellent research in networking. This addresses specifically security, network management, and telecommunication economic services, which are analyzed and prototyped in the context of highly decentralized systems, such as Internet-of-Things IoT networks and Blockchains, by applying, amongst others, Artificial Intelligence AI and other optimization mechanisms. In that field the CSG is committed to an up-to-date teaching curriculum for Bachelor and Master students.

Our focus is on the link between microscopic agent behaviour and the global, emergent properties of socio-economic and socio-technical systems. Blockchain and cryptocurrencies are central to our research, this includes cryptoeconomics, meso- and macro- properties, bigdata blockchain analytics and forensics, design of blockchain-based systems, and the incentives that are present in them.

The interdisciplinary research group investigated the opportunities and risks of digitalization for sustainable development. How can the two major societal transformations, digitalization and sustainability, be aligned? By developing methods and tools to support sustainable development (Sustainability through ICT) and to assess the footprint of digital technologies (Sustainability in ICT), we provided guidance for decision-makers on the path to a sustainable digital society. The ISR Group made decisive contributions to the emerging research field ICT4S (Information and Communication Technologies for Sustainable Development) from 2010 to 2024.

Our group addresses new data science techniques for complex systems that can be modelled as graphs, with a special focus on deep learning for temporal graph data. Our results are published in top theoretical physics journals like Physical Review Letters or Nature Physics and leading computer science venues like SIGKDD, WWW, Learning on Graphs, or ICSE. We practically apply our graph learning methods in projects with researchers from other disciplines, e.g. in software project management, single-cell biology, educational studies, or computational humanities.