Differentiating Through Dual Prices: End-to-End Policy Learning Under Capacity Constraints
End-to-end training is best suited to settings where resources are genuinely scarce and feasibility matters, and leads the delay-adjusted policy value wherever capacity is scarce, once a waiting period carries even a small cost.
M. Haghi, Mahdi Salmani, Nima Kelidari et al.
· 0 citations