Aims: This special session offers a meeting opportunity for academics and industry researchers belonging to the communities of Computational Intelligence, Machine Learning, Experimental Design, Causal Discovery, and Data Mining to discuss new areas of active learning and experimental design, and to bridge the gap between data acquisition or experimentation and model building. The focus is on how active sampling and data acquisition should contribute to the design and modeling of highly intelligent learning systems.

Machine learning prescribes methods and algorithms, which allow a model to learn a behavior from examples. Active learning gathers methods, which select subsets of examples or variables to be used to build a training set for the predictive model. Strategies must be devised to select a subset of examples and variables as small and informative as possible for a task at hand. As a special case, we consider the problem of causal discovery in which one must uncover variables susceptible of influencing a target of interest quantitatively, due to a cause-effect relationship, and check such hypothesis experimentally. Research on incremental experimental design is particularly relevant to this call.

When designing active learning algorithms for real-world data, some specific issues are raised. The main ones are scalability and practicability. Methods must be able to handle high volumes of data, in spaces of possibly high-dimension, and the process for labeling new examples by an expert must be optimized. This includes making "de novo" queries or equivalently for causal systems "manipulating" given variables.

Publication opportunities: Papers should be submitted to IJCNN. We encourage papers that describe applications of active learning in real-world. In the industrial context, the main difficulties met and the original solutions developed, have to be described. Authors of papers accepted in the ECML-ALRA workshop (which do not have any "referenced" proceedings) are also encouraged submit a long version of their paper (up to the maximum number of pages at IJCNN). We are also planning a special topic of JMLR on the theme of experimental design to uncover causal relationships, which will be announced shortly.