I am currently working on my master’s thesis: “Attention-based Multi-Horizon Energy Forecasting”. The goal is to evaluate how far in the future attention-based/transformer models can predict different energy time-series of Germany and in which cases they are worth their complexity compared to common models.