Taught by: Guillaume Hennequin, Kris Jensen, University of Cambridge (July 8, 2021)
Video: Learning What We Know and Knowing What We Learn: Gaussian Process Priors for Neural Data Analysis
Description: Guillaume Hennequin and Kris Jensen, University of Cambridge
Colab notebooks:
- Introduction to FA and GPFA as probabilistic generative models
- Fitting an example data set from a primate reaching task with GPFA
Additional Resources:
- Rasmussen & Williams (2006): The standard textbook for Gaussian processes.
- David Duvenaud’s kernel cookbook: An overview of different covariance functions commonly used for Gaussian processes.
- Rutten et al. (2020): Primary reference for Gaussian process factor analysis with dynamical structure (GPFADS).
- Jensen and Kao et al. (2021): Primary reference for Bayesian GPFA.
- Jensen et al. (2020): Extension of Gaussian process latent variable models to non-Euclidean manifolds.
- Nieh et al. (2021): Demonstration that the hippocampus encodes additional latent structure as well as position in an evidence accumulation task.
