Bounds for Multistage Stochastic Programs Using Supervised Learning Strategies
نویسندگان
چکیده
We propose a generic method for obtaining quickly good upper bounds on the minimal value of a multistage stochastic program. The method is based on the simulation of a feasible decision policy, synthesized by a strategy relying on any scenario tree approximation from stochastic programming and on supervised learning techniques from machine learning.
منابع مشابه
Barycentric Bounds in Stochastic Programming: Theory and Application
The design and analysis of efficient approximation schemes is of fundamental importance in stochastic programming research. Bounding approximations are particularly popular for providing strict error bounds that can be made small by using partitioning techniques. In this article we develop a powerful bounding method for linear multistage stochastic programs with a generalized nonconvex dependen...
متن کاملMachine Learning Solution Methods for Multistage Stochastic Programming
This thesis investigates the following question: Can supervised learning techniques be successfully used for finding better solutions to multistage stochastic programs? A similar question had already been posed in the context of reinforcement learning, and had led to algorithmic and conceptual advances in the field of approximate value function methods over the years (Lagoudakis and Parr, 2003;...
متن کاملScenario-Tree Decomposition: Bounds for Multistage Stochastic Mixed-Integer Programs
Multistage stochastic mixed-integer programming is a powerful modeling paradigm appropriate for many problems involving a sequence of discrete decisions under uncertainty; however, they are difficult to solve without exploiting special structures. We present scenario-tree decomposition to establish bounds for unstructured multistage stochastic mixed-integer programs. Our method decomposes the s...
متن کاملMonotonic bounds in multistage mixed-integer linear stochastic programming: theoretical and numerical results
Multistage stochastic programs bring computational complexity which may increase exponentially in real case problems. For this reason approximation techniques which replace the problem by a simpler one and provide lower and upper bounds to the optimal solution are very useful. In this paper we provide monotonic lower and upper bounds for the optimal objective value of a multistage stochastic pr...
متن کاملBounds and Approximations for Multistage Stochastic Programs
Consider (typically large) multistage stochastic programs, which are defined on scenario trees as the basic data structure. It is well known that the computational complexity of the solution depends on the size of the tree, which itself increases typically exponentially fast with its height, i.e. the number of decision stages. For this reason approximations which replace the problem by a simple...
متن کامل