• Machine learning
• Probabilistic models
• Computational neuroscience
• Information theory

In our group, we focus on developing flexible probabilistic and machine learning methods for modelling and analysing neural activity. We employ deep learning models, such as Transformers, for predicting brain activity. Techniques like copulas, Gaussian processes and normalizing flows are used to describe varying interactions within neural activity and their correlation with external variables. We also work on decomposing matrix and tensor representations of large neural population recordings, which helps in extracting compact patterns for clearer interpretation.