Stefano V. Albrecht

Stefano V. Albrecht

Address

Informatics Forum, 2.06
10 Crichton Street
Edinburgh EH8 9AB
United Kingdom

Contact

ku.ca.de@thcerbla.s
svalbrecht.de

News

About

I am a Lecturer (Assistant Professor) in Artificial Intelligence in the School of Informatics at The University of Edinburgh, where I am affiliated with the Centre for Intelligent Systems and their Applications. I also consult for science and technology organisations, including DARPA and FiveAI. See my CV for more details.

My research interests are in the areas of autonomous agents, multi-agent systems, machine/reinforcement learning, and game theory, with a focus on sequential decision making under uncertainty. The long-term goal of my research is to create intelligent autonomous agents that can interact effectively with other agents to accomplish tasks in complex dynamic environments. This involves a number of challenging problems, such as efficient learning and adaptation in the presence of uncertainty as well as robustness with respect to violations of prior beliefs. The video on the right describes some of my work toward this goal. See publications for more details.

Students

I am looking for talented students who have a keen interest in autonomous systems and want to work on challenging research problems. Top grades, excellent programming skills, and the ability to work independently are essential.

If you need inspiration for research topics, take a look at this recent survey and its section on open problems. Feel free to contact me if you want to discuss research ideas.

I have a fully funded PhD position available for a project in the area of artificial intelligence and cyber security.

You may also consider applying to one of our Centres for Doctoral Training (CDT):

For a list of additional PhD funding, see here.

Teaching

2017/18: Informatics 2D: Reasoning and Agents

Publications

Working papers (under review): Journals: Conferences: Workshops: Magazines:

Software

As a student at TU Darmstadt, I wrote an open-source Matlab toolbox for Support Vector Machines (SVMs). The toolbox supports both inductive and transductive SVMs, multi-class classification, automatic parameter optimisation, and includes a demo function with 2D visualisation. See the manual for more details.
Valid HTML 4.01 Transitional Valid CSS