With a groundbreaking research such as an experience on Deep Learning (DL) for robotics, perception and human poses estimation, adapting DL algorithms, active perception, and how DL models generalized to handle unexpected situations, the Computational Intelligence and Deep Learning research group (CIDL) from AUTH will contribute to specifying, developing and deploying Machine Learning-driven processes and modules of the MAPE-K loop particularly, monitoring, state analysis, learning and adaptation following the uncertainties.