Multitask Support Vector Regression for Solar and Wind Energy Prediction

摘要: Given the impact of renewable sources in overall energy production, accurate predictions are becoming essential, with machine learning a very important tool this context. In many situations, prediction problem can be divided into several tasks, more or less related between them but each its own particularities. Multitask (MTL) aims to exploit structure, training models at same time improve on results achievable either by common model task-specific models. paper, we show how an MTL approach based support vector regression applied photovoltaic and wind energy, problems where tasks defined according different criteria. As shown experimentally three datasets, clearly outperforms specific for least quite competitive energy.