Ihle, Stephan

Stephan Ihle

PhD student, Jan. 2018 to Dec. 2023

Group project, 2015

Research Interests

  • Machine learning
  • Structured neural networks
  • Network plasticity
  • Simulation of small neural networks
  • Nanowire detection

Network Plasticity

Activity induced plasticity
Investigation of stimulation-induced network plasticity. A) A microscopy image of a small directional network containing four nodes. The nodes are the locations where neurons are placed. Neural behavior is recorded at the electrodes. B) Stimulation protocol that was repeated for 60 minures in order to induce learning behavior in the network. C) Examples of spike trains recorded just before stimulation. D) Examples of spike trains recorded just after stimulation. E) A generative recurrent neural network was trained to classify the spike trains before (yellow shaded area) and directly after stimulation (green shading). The further away the value from zero, the more convinced the artificial neural network of a stimulation period occuring in the near past. The stimulation effect cannot be identified 17 h after stimulation.

Network Simulation

simulated network
A simulated network consisting of three nodes. A) The network architecture. Each node consists of ten neurons. For each neuron, a color-coded axon was grown using an inhouse axonal growth model. B) Example spike trains for three nodes. Each vertical line describes an action potential of a neuron in its corresponding node.

Student Projects

Feel free to contact me, if you are interested in any kind of project with me. I am sure we can find a project that suits your interests. I can offer you both theoretical projects (e.g. machine learning, circuit simulation) and practical projects (e.g. cell experiments). If you are interested, please send me your CV and an up to date screenshot from mystudies.

Open projects:

Open projects can be found by contacting me directly or on external pageSiROP (click 'Search Organization' and search for 'lbb').

JavaScript has been disabled in your browser