Antonio Sutera, Arnaud Joly, Vincent François-Valet, Aaron Qiu, Gilles Louppe, Damien Ernst and Pierre Geurts.

In this work, we propose a simple, but yet efficient, method for the problem of connectome inference in calcium imaging data. The proposed algorithm is made of two steps. First, processing the raw signals to detect neural peak activities. Second, inferring the degree of association between neurons from partial correlation statistics. This paper summarizes the methodology that led us to win the Connectomics Challenge, proposes a simplified version of our method and finally discusses our results with respect to other inference methods.

• Contact: a.sutera@ulg.ac.be
• Code: https://github.com/asutera/kaggle-connectomics