TU Berlin

Neural Information ProcessingTheses Projects (Bachelor & Master)

Neuronale Informationsverarbeitung

Page Content

to Navigation

Bachelor and Master Theses

Thank you for your interest in writing your bachelor or master thesis in the field of neural information processing. The focus of possible thesis projects will be closely linked to ongoing research projects of our group. Below you can find open thesis projects and the corresponding contact persons.  If the topic you are interested in is not listed, please contact .

Potentielle Themen

Safe Reinforcement learning
RL agents are typically trained to maximize a reward signal, which must be specified in advance by a human designer. If the reward function isn’t properly designed, the trial-and-error nature of RL in the real world might lead to unacceptable catastrophes. We'll develop safe RL algorithms by transforming the optimization criterion to include some risk measures (Risk-sensitive MDP or Constrained MDP) and deploy our algorithms in the deepmind safety gridworlds and openai safety gym. Write to me if you are interested in deep RL, safe or meta.

Exploring Biologically inspired features for sound event detection in difficult Polyphonic acoustic scenes

Exploring Blind source separation for Sound Event Detection in Moving  Acoustic scenes

Understanding Semantic Segmentation of Sound Events in Binaural Auditory Scenes

Semi-Supervised and Data Augmentation Schemes for Polyphonic Sound event detection in Reverberant  Binaural auditory scenes

Training Neural Networks with evolutionary methods instead of back propagation (MSc)
The standard way for training Neural Networks (NN) is by calculating the gradient of the NN function with respect to the weights/parameters and move along its direction. This method is dominant, mainly due to the fact that the gradient is computationally cheap to calculate. However alternative methods exist. One of this method is the evolutionary approach or gradient free optimization (depending on the community). In this method, one randomly samples various points around the current value of the parameters, and picks a weighted mean of this points as the new point. The advantage of this approach is that it can escape local minima, and that can be applied to cases where the error functions are not differentiable. The goal of the student is to compare these methods in various toy problems.

Comparing different approaches for creating templates out of brain MRIs (MSc)
In MRI studies, it is common to create various templates out of the different groups, before one makes the analysis. These templates are some form of averages, which can be used either as a baseline, or as the objects to be compared between groups. The last decade, a method for calculating averages of images has been developed using either the Wasserstein or the Hellinger-Kantorovich distance. Therefore this method can also be applied for MRIs to create templates. The purpose of this thesis is the comparison of traditional methods for creating templates with these newer approaches.

Training Neural networks that capture the Wasserstein and Hellinger-Kantorovich
distance (MSc)

The purpose of this project is the training of Neural networks that capture the Wasserstein Metric for a specific dataset (MRI or MNIST images). The NNs will be then analyzed and possibly applied to other projects.

Computational modeling of separable resonant circuits controlled by different interneuron types
(Computational Neuroscience, Biophysically Detailed Modeling)

Deficits in Auditory Steady-State Responses in Patients with Schizophrenia in highly realistic Model of Auditory Cortex (Neuropsychiatric Disorder, Computational Neuroscience, Computational Psychiatry)
This is an exciting collaboration with Prof. Salvador Dura-Bernal (SUNY Downstate, www.dura-bernal.org) suited for a Master thesis and will likely result in a publication.

Risk-sensitive deep reinforcement learning

Something related to meta-learning, transfer learning and "towards understanding of neural networks". (Please write me if you are interested)

EEG: measurements, data analysis, study design
In our EEG-lab, we offer a variety of EEG (Electroenzephalogram) related topics like (but not limited to):
Realtime synchronization of data: Using realtime capable equioment (CRIO, NI), incoming measurement data (digital, packeted, UDP) should be "tagged" with absolute timestamps (when the data was collected, not when it arrived), using a reference signal by means of a phase-locked-loop (PLL). After registration the EEG signals should be analyzed (basic methods but implemented in strict RT on an FPGA). Folowing this analysis, the next null-crossing of a likewise extracted stable oscillation should be estimated, send to EEG system and recorded along with the live EEG to check for consistancy. At least good knowledge of "LabView" required.
Source-back-projection: When measuring EEG, neuronal mass signals are collected at sensors (electrodes) which are subsequently used to deduct brain function. However, in many cases, such analysis of signal levels is highly misleading due to the volume conductance of the human head. The electric signals travel through the partly (skull, white matter) highly anisotropic human head, such that a high amplitude at some sensor does not nescessarily has its origin directly below it. The projection from source to sensors can be described by the so called leadfield matrix using estimates. To reconstruct the signal sources, this matrix has to be inverted which leads to a highly ill-posed problem. Some basic knowledge about electric signals and good math skills would be helpfull for a thesis in this field.
EEG studies: possible upon consultation as well as other EEG-related topics


Quick Access

Schnellnavigation zur Seite über Nummerneingabe