From ImageWiki

Jump to: navigation, search

E-mail at the Department of Computer Science: jespersa


Current projects

An experiment in machine learning

An attempt to compete at DTU RoboCup, by applying the LRS pattern matching algorithm as a robot control algorithm for a Scorpion robot. A report on the system, the source code as well as quite short videos of the system in action are available.

The contestant name at DTU RoboCup was [[M�moire]].


Solution for exercise 2.35. (Friday, 15:20. Initial revision: Thursday, Feb 15th, 14:45)

Biologically-inspired machine learning

Functionality of the neocortex (part of the cerebral cortex) has been suggested to be based on a hierarchical bayesian framework: Hierarchical Bayesian inference in the visual cortex.


Key researchers:


Recommended reading

For computer science students with interests like mine:

  • The small book that explains how the brain works: On Intelligence, Jeff Hawkins with Sandra Blakeslee, 2004 - if you think artificial neural networks such as perceptrons doesn't really deserve the name "neural", but still think that much can be learned from the way brains work.
  • A more in-depth article, Hierarchical Bayesian inference in the visual cortex, building on the established Bayesian belief propagation (just like in On Intelligence) and particle filtering algorithms.
  • Cognitive Neuroscience, Gazzaniga et al. 2002 - on the biological foundations of cognitive phenomena - textbook used on the cognitive psychology course at the University of Copenhagen.

To watch:

Studies abroad

Cf. Studieophold i udlandet

Personal tools