2 Postdoc positions in Computational Neuroscience
Veröffentlicht:
13 April 2025Pensum:
100%- Arbeitsort:Bern
Job-Zusammenfassung
Die Universität Bern sucht 2 Postdoc-Positionen in der Computational Neuroscience. Eine spannende Gelegenheit zur Forschung in einem dynamischen Umfeld.
Aufgaben
- Entwicklung eines Modells für kortikale Selbstaufmerksamkeit.
- Anwendung neuronaler Prinzipien auf langzeitliche Verarbeitung.
- Zusammenarbeit mit anderen Labors in KI und neuromorpher Technik.
Fähigkeiten
- PhD in Computational Neuroscience, Machine Learning und Mathematik erforderlich.
- Starke Kenntnisse in mathematischen Modellen.
- Erfahrung mit neuronalen Netzwerken und KI.
Ist das hilfreich?
2 Postdoc positions in Computational Neuroscience
Published 11 April 2025 Workplace Bern, Bern region, Switzerland CategoryComputer Science
2 Postdoc positions in Computational Neuroscience
100%
Department of Physiology
01.06.2025 or by agreement
2-year contract, with the possibility of extension for a third year.
The first position focuses on developing a computational model of cortical self-attention( Publication- Granier-2025Multiheads ). Building on our current work and an ongoing collaboration with The Virtual Brain ( EBRAINS ), we are seeking to implement neuronal self-attention mechanisms in thalamo-cortical circuits. The model is inspired by transformer-type architectures but remains consistent with experimentally observed cortical connectivity patterns. It will be trained on cognitive tasks while being constrained by human cortical recordings.), we are seeking to implement neuronal self-attention mechanisms in thalamo-cortical circuits. The model is inspired by transformer-type architectures but remains consistent with experimentally observed cortical connectivity patterns. It will be trained on cognitive tasks while being constrained by human cortical recordings.
The second position will build on the Neuronal Least-Action principle ( Publication- previewed-preprints ) and its extension to long-term temporal processing and spike-based activity. The framework provides a rigorous description of neuronal dynamics in cortical networks, along with gradient-based synaptic learning rules. It will be applied to integrate-and-fire neurons with multiple intrinsic time constants. The project also connects to implementations in spike-based neuromorphic hardware.
Requirements
PhD with a strong background in computational neuroscience, machine learning and mathematics.
We offer
- Interdisciplinary & international research environment
- Interaction with experimental neuroscience and close collaboration with other labs in artificial intelligence and neuromorphic engineering.
If you are interested, please send one .pdf document including your CV, a statement of research interests, a list of publications and the name and contact details of two referees to .ch and latest by April 28, 2025.
