ICNN-enhanced 2SP uses architecturally convex neural networks to enable exact LP embedding of recourse surrogates, replacing MIP formulations and yielding up to 100x speedups on benchmark problems.
arXiv preprint arXiv:2112.00874 , year=
2 Pith papers cite this work. Polarity classification is still indexing.
2
Pith papers citing it
fields
math.OC 2representative citing papers
Introduces FR-NDPs as risk-averse multi-stage stochastic programs with SDDP convergence conditions, demonstrated on SAV capacity expansion and SAV-BRT integration in a Manhattan network.
citing papers explorer
-
ICNN-enhanced 2SP: Leveraging input convex neural networks for solving two-stage stochastic programming
ICNN-enhanced 2SP uses architecturally convex neural networks to enable exact LP embedding of recourse surrogates, replacing MIP formulations and yielding up to 100x speedups on benchmark problems.
-
Flexible and Reliable Network Design for Emerging Transportation Services: Multi-stage Stochastic Programming Approach
Introduces FR-NDPs as risk-averse multi-stage stochastic programs with SDDP convergence conditions, demonstrated on SAV capacity expansion and SAV-BRT integration in a Manhattan network.