Preprint
Aug 2026
Hard Constraints, Smooth Gradients: Learning Feasible Inventory Policies via Differentiable Projection
This work shows that DRL can deliver economically significant savings in sequential decision problems with interdependent hard constraints, which are widespread in practice and where planning is hardest: in tightly capacitated systems with high demand variability.
P. Helm, John C. Doerr, Joren Gijsbrechts et al.
· 0 citations