Transformer-based Stagewise Decomposition for Large-Scale Multistage Stochastic Optimization

arXiv:2404.02583v1 Announce Type: new
Abstract: Solving large-scale multistage stochastic programming (MSP) problems poses a significant challenge as commonly used stagewise decomposition algorithms, including stochastic dual dynamic programming (SDDP), face growing time complexity as the subproblem size and problem count increase. Traditional approaches approximate the value functions as piecewise linear convex functions by incrementally accumulating subgradient cutting planes from the primal and dual solutions of stagewise subproblems. Recognizing these limitations, we introduce TranSDDP, a novel Transformer-based stagewise decomposition algorithm. This innovative …

abstract algorithms arxiv challenge complexity count cs.lg dynamic face functions linear msp optimization programming scale stochastic transformer type value