Members

EPFL Labs

We study and design collaboration technologies that lead to better learning and more efficient problem-solving.

Research in our group focuses on efficient peer-2-peer systems, trust management, large-scale semantic interoperability, and self-organization in decentralized information systems.

The DHLAB ambitions to form a new generation of young researchers – “digital humanists”. It develops innovative educational programs and didactic technologies to progress towards this goal.

Our main activity is system design for miniature autonomous mobile robots. This field includes a large spectrum of disciplines, spanning from Artificial Intelligence to mechatronics, from navigation techniques to digital electronics, from sensors to actuators technologies. Our research is based on strong competences in digital electronics and system integration.

The interdisciplinary research carries out in the React Group aims at handling real-time coordination and distributed interaction challenges using systems engineering methodologies. 

The Instant-Lab conducts research in micromechanical design. The key application field on which the research is focused is horology. A wider spectrum of applications is also concerned including: medtech applications, precision robotics, space applications, astrophysics applications and instrumentation for accelerator facilities.

Jonathan Ledgard is director of the Afrotech lab at EPFL. Afrotech seeks to pioneer massively scaleable high technologies into and out of Africa [including distribution of advanced knowledge on digital platforms, robotics, solar, and complexity thinking for future cities]. Prior to joining EPFL Jonathan was a foreign political and war correspondent for The Economist. He joined the newspaper in 1994. Since 2004 he was Africa correspondent for the paper, with a reporting focus on risk and technology issues. He has published two acclaimed novels, Giraffe (2006) and Submergence (2012).
Please join us for a talk by Jonathan Ledgard

 

The laboratory is dedicated to the development and application of quantum simulations to understand, predict, and design the properties of complex materials and devices, with a core effort in energy and nanotechnology. We are very actively engaged in the development and dissemination of open-source software for quantum simulations, with more than 20 schools and workshops in the developed and developing world, and the first online lecture series on materials modelling (ocw.mit.edu/3-320S05, taught simultaneously at MIT, U. of Cambridge, and NUS/NTU Singapore). Prof Marzari spent 2008 as the RP Feynman Professor of Nanosciences at the University of Addis Ababa, teaching computational materials science.

 

Our group develops computational methods to analyze human behavior in social media (in particular in social video) and a variety of ubicomp settings, including small groups at work in multisensor spaces, and populations of smartphone users in urban environments.

Research Projects

We built a MOOC-BOOK player which links text-book content with MOOC videos automatically to help them to study in groups.

In MOOCollab, we bring together the affordances of collaboration and study groups to the MOOC context. MOOCollab is a collaborative environment which enables a group (4 to 5 individuals) to watch MOOC videos together and solve exercises in a collaborative fashion.

The goal of the three years long project is the creation of new mechanisms for supporting users in the evaluation of Web content credibility and improvement of the quality of Web content credibility evaluation. Our society increasingly relies on the vast amount of information available on the Internet (and the WWW in particular). As a consequence, the issue of judging or evaluating the credibility of this information becomes of crucial importance. Thus, algorithms and mechanisms for the assessment of credibility of Web content can have a great, positive social impact.

Is it possible to identify the memorable events of a user’s life, with few if no user feedback? MemorySense seeks an answer to this question by carefully intertwining sensor data and heterogeneous data sources  obtained from mobile phones, in a quest to better understand how human memories work. The project  covers both theoretical challenges (e.g., learning and forgetting models) and engineering aspects (e.g., mobile data gathering, UI design in Android, etc.). MemorySense is a joint effort between EPFL and Samsung Research (SISA R and D Center).

The Go-Lab Project opens up remote science laboratories and their online models (online labs) for the large-scale use in education. Its technical framework – the Go-Lab Portal – offers students the opportunity to perform personalized scientific experiments with online labs, whereas teachers may enrich their classroom activities with demonstrations and disseminate best practices in a web-based pedagogic community.

This project deals with the extensions of the Swiss e-Infrastructure for self-organized and competence-oriented learning environments. In view of the vast amount of contents, data resources and tools, learning becomes increasingly complex. Students are expected to get orientation on their own, and to determine and elaborate adequate, individual learning methods.

This project aims at developing and disseminating Massive Open Online Labs (MOOLs) methodologies and infrastructures in cooperation with Tunisia and Morocco, with a special focus on providing remote labs related to the production of green energy.

Conversational video (in the form of video blogs or video reviews) is a hugely popular genre in platforms like YouTube. VlogSense leverages research in social media and multimedia analysis to characterize nonverbal and verbal behavior of social video users, with the goal of building automated tools for discovery of and interaction with online video content.