Predicting Electricity Consumption with Random Walks on Gaussian Processes
This work addresses the challenge of high computational costs in time-series forecasting for energy stakeholders, but it is incremental as it builds on existing Gaussian Process methods.
The paper tackled the problem of forecasting electricity consumption with scarce data by developing an algorithm called Domino that uses random walks on Gaussian Processes to reduce training costs, achieving computational savings in experiments.
We consider time-series forecasting problems where data is scarce, difficult to gather, or induces a prohibitive computational cost. As a first attempt, we focus on short-term electricity consumption in France, which is of strategic importance for energy suppliers and public stakeholders. The complexity of this problem and the many levels of geospatial granularity motivate the use of an ensemble of Gaussian Processes (GPs). Whilst GPs are remarkable predictors, they are computationally expensive to train, which calls for a frugal few-shot learning approach. By taking into account performance on GPs trained on a dataset and designing a random walk on these, we mitigate the training cost of our entire Bayesian decision-making procedure. We introduce our algorithm called \textsc{Domino} (ranDOM walk on gaussIaN prOcesses) and present numerical experiments to support its merits.