research

in motion

You are here: Home Research Semester & master theses

Semester & master theses


If you are interested in doing your Semester Project or Master Thesis in the Stoop group, please contact Prof. Ruedi Stoop. You can come with your own idea for the project, or we can also propose you a project idea which will be linked to the research directions pursued in the Stoop group.

Semester project: Analysis of 2D-CA or Asynchronous CA

This Semester Project aims to study the properties of two-dimensional cellular automata (CA) and/or of asynchronous CA.

Cellular automata (CA) have a long and intricate history. Already von Neumann studied in the 1940s biologically motivated CA in order to design self-replicating artificial systems. A specific CA, the ,,game of life” proposed by Conway in 1970 rose to special prominence.

Our main interest with CA lies in the fact that they are among the oldest models of natural computation : They are massively parallel, memory and processing are not separated, interactions are local and they can exhibit self-organizing properties as living matter does. If programmed accordingly, they further show physical properties as reversibility or conservation laws.

In this project we want to study 2D-CA and asynchronously updating CA with tools from dynamical systems theory and statistical mechanics.

Requirements:

  • Bachelor degree in physics or a related field.
  • Statistical physics.
  • Mathematically and algorithmically inclined student.
  • Good programming skills in C, Mathematica, Mathlab or equivalent.

 

Embedding of the Hopf cochlea into an cortico-cochlear feedback loop

The mammalian auditory system extracts from the arriving soundwaves rich portraits of the auditory environment. To a large extent, this ability is the consequence of the nonlinear signal processing in the cochlea, where the local amplification profiles have been shown to originate from systems close to Hopf bifurcations. Starting from biophysical principles, we have developed and implemented an analog electronic Hopf-cochlea displaying mammalian hearing characteristics (patented). Using this novel sensor, we aim at implementing in this project an autonomous hearing system, supported by an auditory cortico-cochlear feedback loop. In particular, we investigate the role of cochlear processing in assisting a listener to discriminate between different auditory objects in a cocktail-party environment and how much this task can be improved by an efferent tuning of the Hopf amplifiers.

cochleaFilter

Various projects are available in the following fields:

  • Analog circuit implementation
  • Neural network implementation (software)
  • Signal processing
  • Theoretical development of a quality measure

 

Requirements:

Mathematically inclined student, pleasure in analog electronics (audio).

 

Assets:

  • German language
  • Nonlinear electronic circuits
  • Programming skills in C, Mathematica, LabView, Matlab
  • Statistical physics

 

Quest for nonlinear dynamics footprints in bioinformatics and system biology

As systems and their interactions in the fields of bioinformatics and system biology are generally nonlinear, the applications of tools of nonlinear dynamics to this area is a rapidly growing field, but still in its infancy. The Stoop group at the Institute of Neuroinformatics UZH/ETHZ deals with related research questions since more than fifteen years.

In this master project, the quest will be to reveal in biophysics data the footprints of lowdimensional nonlinear dynamics processes, which are universal features of broad classes of nonlinear systems. From these footprints, detailed conclusions can be drawn on the nature of the underlying biological systems. The methods that will be employed are from the area of statistical mechanics and artificial intelligence, such as clustering algorithms and genetic programming. We will need to adapt these approaches in order to become optimized in view of this application. The emerging computational tools and concepts will, however, reach beyond this application: They are expected to become a fundamental part of the bioinformatics and system biology approach.

shrimpsNumerical

Requirements:

  • Bachelor degree in physics or a related field.
  • Mathematically and algorithmically inclined student.
  • Good programming skills in C, Mathematica, Mathlab or equivalent.

Assets:

  • German language.
  • ETH-courses “Theorie, Programmierung und Simulation neuronaler Netze”, “Complex Systems: Berechenbares Chaos in dynamischen Systemen”, and "Dynamische Systeme in der Biologie".
  • Statistical physics.

 

Perception by means of clustering and visual scene analysis

This Master Project (ev. Semester Project) targets the analogies between human visual perception vs. recent statistical physics approaches to perception. It is composed of an experimental and a theoretical part, resulting in a truly interdisciplinary research project.

When searching for objects in a unknown environment, the human visual system is not done based on random strategies. We are able to screen scenes very fast and to almost immediately extract the objects we are interested in. In the field of machine learning, statistical mechanics based algorithms have recently been developed that begin to challenge human perception (e.g., Sequential Superparamagnetic Clustering Algorithm and Hebbian Learning Based Self-Organizing Neural Networks).

Using the eye as the window into the brain, we would like to learn to what extent the analogy of human visual perception and the developed artificial perception strategies hold. Can we learn from the visual system how to improve the artificial intelligence approaches? Can the comparison with the artificial intelligence approach provide conclusions on the nature of the neural systems guiding human visual perception?

In the experimental part, a state-of-the-art eye-tracking system will be used to collect in a non-invasive way data from human subjects. The visual search task will be defined and measured in experiments. In the theoretical part, existing methods and algorithms will be analyzed and adapted in order to come as close to the way the human system solves these problem.

The whole topic is terra nova, with ample space for innovative, independent approaches by the student.

 

Requirements:

  • Bachelor degree in physics or a related field.
  • Statistical physics.
  • Mathematically and algorithmically inclined student.
  • Good programming skills in C, Mathematica, Mathlab or equivalent.

 

Assets:

  • German language.
  • ETH-courses "Theorie, Programmierung und Simulation neuronaler Netze", "Complex Systems: Berechenbares Chaos in dynamischen Systemen", and "Dynamische Systeme in der Biologie".
  • Experience in psychophysical experiments.