============
This is an implementation of the multi-view representation learning and trace regression models for mapping individual differences using multimdodal fMRI data as described in our paper:
A. Sellami et al. Mapping-individual-differences-in-cortical-architecture-using-multi-view-representation-learning, IJCNN conference (2020)
We proposed a multimodal deep autoencoder (MDAE) framework that allows combining the activation- and connectivity-based fMRI protocols to identify markers of individual differences. MAEs are trainable neural network models for unsupervised learning and dimensionality reduction.
This research work is carried out jointly with Qarma (Machine Learning Team), LIS laboratory and the Banco (Neural Bases of Communication) team at Institute of Neuroscience of Timone (Institut de Neurosciences de la Timone a.k.a. INT ). This project is funded by the Institute of Language, Communication and the Brain (ILCB)