GGS Colloquium Speaker: Matthias Schubert
“REINFORCEMENT LEARNING WITH SPATIAL APPLICATIONS”
Smart cities offer more and more real-time information
provided by sensor networks and traffic cameras. This information can be
very valuable for transportation planing and analysing user behaviour.
For instance, knowing which parking spots are currently available close
to my destination is very valuable in order to reduce the travel time
and thus, maximize the resource usage and minimize the traffic load. The
future development of this information is usually uncertain (i.e.
non-deterministic). However, algorithms for routing applications should
consider that new information will become available during travelling
along the computed path. In order to exploit the provided information to
a full extend, it is not sufficient to compute a static route or travel
plan because the optimality of the plan might degrade as the state of
the environment might consistently change.
In order to plan transportation and understand observed trajectories in
smart environments, it is necessary to compute action policies instead
of static routes. A policy provides the most promising action for all
situations and in particular, the encountered situations. Analogously,
it makes sense to understand human behaviour based on the sequence of
decisions in the encountered situations. To compute and analyse
policies, the field of reinforcement learning already provides a rich
set of tools.
Matthias Schubert’s Bio:
Matthias Schubert is a professor for Computer Science at the LMU
Munich. He is one of the founders of the Data Science Lab @LMU Munich
and a member of the Munich Competence Center for Machine Learning. His
research interests comprise spatial information systems, representation
learning and artificial intelligence in non-deterministic environments.