Machine learning for neuronal activity

Decoding neuronal activity is at the intersection between machine learning, statistics and neuroscience. The objective of this project is to develop and implement methods that can be used to learn from very large amounts of calcium data; including source extraction, dimensionality reconstruction, relation to external stimuli and clustering based on functionality of different brain areas. Methods from Deep Learning and Graphical Models will be used.

There are a lot of existing methods to extract signal sources from calcium data and behaviour, and relate them to external stimuli. This project focuses on translating the power of deep learning to this area in order to analyse the data in an end-to-end fashion.