Abstract:We study two classic variants of block-structured integer programming. Two-stage stochastic programs are integer programs of the form $\{A_i \mathbf{x} + D_i \mathbf{y}_i = \mathbf{b}_i\textrm{ for all }i=1,\ldots,n\}$, where $A_i$ and $D_i$ are bounded-size matrices. On the other hand, $n$-fold programs are integer programs of the form $\{{\sum_{i=1}^n C_i\mathbf{y}_i=\mathbf{a}} \textrm{ and } D_i\mathbf{y}_i=\mathbf{b}_i\textrm{ for all }i=1,\ldots,n\}$, where again $C_i$ and $D_i$ are bounded-size matrices. It is known that solving these kind of programs is fixed-parameter tractable when parameterized by the maximum dimension among the relevant matrices $A_i,C_i,D_i$ and the maximum absolute value of any entry appearing in the constraint matrix. We show that the parameterized tractability results for two-stage stochastic and $n$-fold programs persist even when one allows large entries in the global part of the program. More precisely, we prove that: - The feasibility problem for two-stage stochastic programs is fixed-parameter tractable when parameterized by the dimensions of matrices $A_i,D_i$ and by the maximum absolute value of the entries of matrices $D_i$. That is, we allow matrices $A_i$ to have arbitrarily large entries. - The linear optimization problem for $n$-fold integer programs that are uniform -- all matrices $C_i$ are equal -- is fixed-parameter tractable when parameterized by the dimensions of matrices $C_i$ and $D_i$ and by the maximum absolute value of the entries of matrices $D_i$. That is, we require that $C_i=C$ for all $i=1,\ldots,n$, but we allow $C$ to have arbitrarily large entries. In the second result, the uniformity assumption is necessary; otherwise the problem is $\mathsf{NP}$-hard already when the parameters take constant values. Both our algorithms are weakly polynomial: the running time is measured in the total bitsize of the input.

Stochastic programming with integer variables

Multistage Robust Mixed-Integer Optimization under Endogenous Uncertainty

An integer programming approach for linear programs with probabilistic constraints

Introduction to Stochastic Programming

Parameterized algorithms for block-structured integer programs with large entries

Solving Nonsmooth Nonconvex Compound Stochastic Programs with Applications to Risk Measure Minimization

Stochastic Constraint Programming

Algorithms for Stochastic Mixed-Integer Programming Models

Solving a class of two-stage stochastic nonlinear integer programs using value functions

Existence of Solutions in Bi-level Stochastic Linear Programming with Integer Variables

Weak Continuity of Risk Functionals with Applications to Stochastic Programming

Solving Two-Stage Stochastic Programs with Endogenous Uncertainty via Random Variable Transformation

Chapter VIII Stochastic programming

On the ReLU Lagrangian Cuts for Stochastic Mixed Integer Programming

Solving Lagrangian variational inequalities with applications to stochastic programming

An Introductory Tutorial on Stochastic Linear Programming Models

Distributionally Risk-Receptive and Robust Multistage Stochastic Integer Programs and Interdiction Models

On Risk-Averse Stochastic Semidefinite Programs with Continuous Recourse

Block-Structured Integer and Linear Programming in Strongly Polynomial and Near Linear Time

A Risk-Aversion Approach for the Multiobjective Stochastic Programming Problem

Integer Programming Approaches for Distributionally Robust Chance Constraints with Adjustable Risks