Computational Intelligence technologies have made great progress in recent decades. Real world environments produce large-scale, high-dimensional, multi-modal, sequential and ambiguous data. Since many real world problems are not considered to be well-posed mathematically, attempts of analytic approaches to find solutions met some difficulties. For dealing with such complex data, various techniques are required such as visualization by clustering of multi-modal and sequential data, automatic feature extraction by representation learning, acquisition of comprehensible knowledge from learning results and so on. Driven by such motivation, emerging computational intelligence approaches have been proposed in the soft-computing areas like artificial neural networks, evolutionary computation and fuzzy theories, and many of these innovative technologies are now becoming popular in the field of computer science such as pattern recognition, combinatorial optimization problems etc. The advanced technology of the recent successes is Deep Learning, with the advancement of computer hardware providing high performance computing like GPGPU environments. According to the brisk activities, many researchers also have been able to challenge solving industrial problems such as the control system of industrial robots, analysis of medical database, etc. We discuss in this session the computational intelligence technologies for learning real world complex data, which will make an explicit or implicit knowledge to the real world problems that prior technologies cannot provide satisfactory solutions.