🔥 Replication of various papers, to which I also add a pdf file with comments both on the paper and on the figures.
🔥 Looking for feedback: both on the efficiency of the implementation and the coding style. These re-implementations are made to be criticized and learn from it!
- The replications are done either in julia or python.
- The goal for me is not to criticize the author work, which I always assume to be of high quality. Indeed, as a student I still have a lot to learn before being able to argue on the implementation of a model. Nonetheless, I will add a description of how hard I found it to replicate the work, from a student perspective.
- Sometimes I will focus on the main figures.
- Additional explorations may be proposed at the end of the work.
Most papers will be either machine learning or computational neuroscience papers.
The Manifest and Project config files are used to obtain the same environment as mine in Julia. Note that since these work are exploratory I used their most of the library I know in Julia, and some might not be needed for some implementations. The implementations are runned on the GPU when possible/useful.
- Boucheny, C., Brunel, N. & Arleo, A. A Continuous Attractor Network Model Without Recurrent Excitation: Maintenance and Integration in the Head Direction Cell System. J Comput Neurosci 18, 205–227 (2005). https://doi.org/10.1007/s10827-005-6559-y