Benjamin Schäfer

Current scoring rules may lead to overconfident forecasts that misrepresent uncertainty, risking poor decision-making in energy markets.

On-policy deep reinforcement learning agents can significantly outperform off-policy methods in optimizing energy management while providing clear insights into their decision-making.

TSFMs can achieve competitive forecasting performance in critical infrastructure applications while also providing interpretable explanations that align with established domain knowledge.

Foundation models don't always win: task-specific models can rival or even beat them in electricity price forecasting, especially with clever feature engineering or transfer learning.