Abstract
We dealt with the challenge of cognitive brain state prediction as proposed by the Pittsburgh Brain Activity Interpretation Competition (PBAIC) 2007. The problem was decomposed in many subsequent steps: pre-processing, feature selection, learning model selection, model training, and post-processing. We investigated the steps combining unsupervised and supervised learning techniques and assessed the most effective tech-
nique to permorm each of them. The final predictions have been produced by using a mixture of different learning models: k-means, recurrent neural networks, gaussian process regression, iterated conditional mean.