Statistical Validation Methods for Complex Systems
Complex Systems Science represents the benchmark framework to analyze a huge variety of interacting systems in a broad spectrum of disciplines, ranging from Biology and Climate Science to Economics and the Social Sciences.
In recent years, many of such disciplines have witnessed a dramatic increase in the availability of data recorded from the activity of their systems of interest. This is unquestionably a positive development, which however poses a fundamental challenge to empirical research, as the more data become available, the more it becomes difficult to identify the relevant interactions in a complex system.
Purely data-driven approaches can be easily misled by an overabundance of data, as highlighted, for example, by the spectacular failure of the Google Flu Trends predictions in 2014. This, in turn, calls for the development of sophisticated statistical techniques capable of extracting meaningful information and reliable predictions from the data deluge we are exposed to.
The goal of this satellite workshop is to bring together experts and practitioners from all areas of Complexity Science whose research is focused on the development of statistical methodologies aimed at identifying and validating the relevant interactions in complex systems. The workshop will bring together researchers from all areas of theoretical and applied Complexity Science, covering subjects such as time series filtering, link validation and prediction in networks, machine learning, and information theory, with the aim of laying the foundations for a shared research agenda.
Submission & Important dates
Please submit a one page abstract by email to firstname.lastname@example.org no later than June 15, 2018
Notification of acceptance: July 1, 2018
Tomaso Aste (University College London, UK)
Why Artificial Intelligence is not effective for real complex systems and how can we fix it?
Ginestra Bianconi (Queen Mary University London, UK)
Guido Caldarelli (IMT Lucca, Italy)
Maximum entropy validation and reconstruction in complex networks
Rosario Mantegna (University of Palermo, Italy)
Statistically validated trading networks in stock markets
Final program coming soon... (half-day format)