Decision-Focused Learning: Through the Lens of Learning to Rank

Evolutionary Scheduling of University Activities Based on Consumption Forecasts to Minimise Electricity Costs

“…Table II provides the accuracy scores achieved with the final model configuration. The accuracy is assessed with the MASE, which is defined in (6).…”

“…They show that the evolutionary scheduling method obtains better costs and presents a trade-off with emission cost or/and reliability. Mandi et al [6] views an energy-cost aware scheduling problem as a learning to rank (LTR) problem. The proposed solution uses surrogate loss functions to cache feasible solutions.…”