Machine Learning Applications for Complex Business Challenges

Over the past decade, data-driven computational techniques have matured to the point where they can now provide effective solutions to complex business challenges. Machine learning, data mining, and autonomous agents, are some of the most prominent techniques in this field, and their applications range from prediction tasks to orchestrating complex business processes. This masterclass explores ground-breaking applications of machine learning in business settings, and highlights their benefits and associated challenges.

Machine learning applications for complex business challenges (Wolf Ketter)
In his opening talk, Prof. Wolf Ketter of the Erasmus Centre for Future Energy Business, and director of the Learning Agent research group at Erasmus (LARGE) showcases a series of machine learning-based research projects that are currently undertaken at Erasmus University. These projects tackle business challenges ranging from high-speed auctions to coordinated e-vehicle charging, where machine learning can effectively replace or support human decision-making.

Causal discovery from big data: mission (im)possible? (Tom Heskes)
Discovering causal relations from data lies at the heart of most scientific research today. In apparent contradiction with the adagio "correlation does not imply causation", recent theoretical insights indicate that such causal knowledge can also be derived from purely observational data, instead of only from controlled experimentation. In the "big data" era, such observational data is abundant and being able to actually derive causal relationships from very large data sets would open up a wealth of opportunities for improving business, science, government, and healthcare.

In this talk, I will give a gentle introduction to the basic principles behind mainstream causal paradigms, discuss the challenges when applying those to big data, and present our own research line towards more robust, informative, and realistic estimates of causal mechanisms.

WindML - A Framework for Data Mining in Wind Power Time Series (Oliver Kramer)
Wind energy is playing an increasingly important part for ecologically friendly power supply. The fast growing infrastructure of wind turbines can be seen as large sensor system that screens the wind energy at a high temporal and spatial resolution. The resulting databases consist of huge amounts of wind energy time series data that can be used for prediction, controlling, and planning purposes. In this work, we describe WindML, a Python-based framework for wind energy related machine learning approaches. The main objective of WindML is the continuous development of tools that address important challenges induced by the growing wind energy information infrastructures. Various examples that demonstrate typical use cases are introduced and related research questions are discussed. The different modules of WindML reach from standard machine learning algorithms to advanced techniques for handling missing data and monitoring high-dimensional time series.

Machine Learning Research that Matters: The case of information acquisition (Maytal Saar-Tsechansky)
In this talk I will argue for the growing importance for machine learning research of considering the particular context in which machine learning is applied and the objectives of systems of which machine learning techniques are components. Drawing from my own research on information acquisition, I will demonstrate the practical benefits from doing so, and how the naïve application of general-purpose machine learning, in the absence of appropriate solutions, not only does not benefit a problem but can rather undermine it. Identifying context-specific research challenges with meaningful, real-world implications, and conducting the research to address these problems present opportunities for a viable and impactful machine learning research community in business schools and engineering.