📚 Assignment

Author

Affiliation

EURECOM

For this course, you will have the opportunity to work on a final assignment.
The assignment will be more research-oriented and consist of reproducing a paper.
- You will have to choose a paper from a list and reproduce (some) the results.
- You will have to write a short report 4explaining the paper and the results.
- You will have to submit the report and the code.
The final list of papers will be available around the middle of the course, with a deadline around the end of the course.
The assignment can be done in group (2 people max) but the report is individual.
The assignment will be graded based on the quality of the report, the quality of the code, and the quality of the results. It will count for 40% of the final grade.

🎯 Objectives

Understand a research paper in the field of (probabilistic) machine learning.
Be able to apply the basics of the course to reproduce the results.
Be able to go from equations to code in a more complex setting than the labs.
Be able to write a short report explaining the paper, the results, and the code.

A short report (4-5 pages max) with the NeurIPS format.
A link to a GitHub or GitLab repository with the code and the results (ideally, a Jupyter notebook).

Weight Uncertainty in Neural Networks. Blundell et al. 2015.
Bayesian Learning via Stochastic Gradient Langevin Dynamics. Welling and Teh. 2011.
Sparse Gaussian Processes using Pseudo-inputs. Snelson and Ghahramani. 2006.
Dropout as a Bayesian Approximation: Representing Model Uncertainty in Deep Learning. Gal and Ghahramani. 2016.
Stochastic gradient Hamiltonian Monte Carlo. Chen et al. 2014.
Cyclic Stochastic Gradient MCMC for Bayesian Deep Learning. Zhang et al. 2019.
Handling sparsity via the Horseshoe prior. Carvalho et al. 2009.
Variational Dropout and the Local Reparameterization Trick. Kingma et al. 2015.
Bayesian classification with Gaussian processes. Williams and Barber. 1998.