Narrow Results

Real-time decision making has acquired increasing interest as a means to efficiently operating complex systems. The main challenge in achieving real-time decision making is to understand how to develop next generation optimization procedures that can work efficiently using: (i) real data coming from a large complex dynamical system, (ii) simulation models available that reproduce the system dynamics. While this paper focuses on a different problem with respect to the literature in RL, the methods proposed in this paper can be used as a support in a sequential setting as well. The result of this work is the new Generalized Ordinal Learning Framework (GOLF) that utilizes simulated data interpreting them as low accuracy information to be intelligently collected offline and utilized online once the scenario is revealed to the user. GOLF supports real-time decision making on complex dynamical systems once a specific scenario is realized. We show preliminary results of the proposed techniques that motivate the authors in further pursuing the presented ideas.

One of the issues associated with analysing data coming from multiple sources is that the data streams can have markedly different spatial, temporal and accuracy characteristics. For example, in air quality monitoring we may wish to combine data from a reference sensor that provides relatively accurate hourly averages with that from a low cost sensor that provides relatively inaccurate averages over finer temporal resolutions. In this paper, we discuss algorithms for analysing multi-fidelity data sets that use the high accuracy data to allow a characterisation of the low accuracy measurement systems to be made. We illustrate the approaches on data simulating air quality measurements in co-location studies in which a reference sensor is used to calibrate a number of other sensors. In particular, we discuss approaches that can be applied in cases where sensors are outputting averages over different time intervals.