TACC2008, Budapest, July 7-9 2008 Tutorial: Self-Preservation and Reactive Search Roberto Battiti and Mauro Brunato (University of Trento) Abstract: A distributed architecture based on autonomic elements and aggregated services can be the target of disruptive actions both from the outside (identity spoofing, denial of service) and from the inside (malicious and selfish participants). Preventing such attacks in a highly dynamic environment requires fast detecton and ready action. In this context, pattern recognition and learning techniques are extremely useful. Reactive Search advocates the integration of sub-symbolic machine learning techniques into search heuristics for solving complex optimization problems. The word "reactive" hints at a ready response to events during the search through an internal online feedback loop for the self-tuning of critical parameters. Methodologies of interest for Reactive Search include machine learning and statistics, in particular reinforcement learning, active or query learning, neural networks, and meta-heuristics.