
Optimal Solution Predictions for Mixed Integer Programs
Mixed Integer Programming (MIP) is one of the most widely used modeling ...
read it

Solving Mixed Integer Programs Using Neural Networks
Mixed Integer Programming (MIP) solvers rely on an array of sophisticate...
read it

Generating Hard Instances for Robust Combinatorial Optimization
While research in robust optimization has attracted considerable interes...
read it

Learning Optimal Temperature Region for Solving Mixed Integer Functional DCOPs
Distributed Constraint Optimization Problems (DCOPs) are an important fr...
read it

Online MixedInteger Optimization in Milliseconds
We propose a method to solve online mixedinteger optimization (MIO) pro...
read it

Learning a Large Neighborhood Search Algorithm for Mixed Integer Programs
Large Neighborhood Search (LNS) is a combinatorial optimization heuristi...
read it

Learning Efficient Search Approximation in Mixed Integer Branch and Bound
In line with the growing trend of using machine learning to improve solv...
read it
Learning PseudoBackdoors for Mixed Integer Programs
We propose a machine learning approach for quickly solving Mixed Integer Programs (MIP) by learning to prioritize a set of decision variables, which we call pseudobackdoors, for branching that results in faster solution times. Learningbased approaches have seen success in the area of solving combinatorial optimization problems by being able to flexibly leverage common structures in a given distribution of problems. Our approach takes inspiration from the concept of strong backdoors, which corresponds to a small set of variables such that only branching on these variables yields an optimal integral solution and a proof of optimality. Our notion of pseudobackdoors corresponds to a small set of variables such that only branching on them leads to faster solve time (which can be solver dependent). A key advantage of pseudobackdoors over strong backdoors is that they are much amenable to datadriven identification or prediction. Our proposed method learns to estimate the solver performance of a proposed pseudobackdoor, using a labeled dataset collected on a set of training MIP instances. This model can then be used to identify highquality pseudobackdoors on new MIP instances from the same distribution. We evaluate our method on the generalized independent set problems and find that our approach can efficiently identify highquality pseudobackdoors. In addition, we compare our learned approach against Gurobi, a stateoftheart MIP solver, demonstrating that our method can be used to improve solver performance.
READ FULL TEXT
Comments
There are no comments yet.