Communications of the ACM
ML Speeds Up Vehicle Routing
Researchers used a machine-learning-augmented method to speed up the solutions to vehicle routing problems for large sets of cities.
Credit: Shutterstock
Researchers at the Massachusetts Institute of Technology (MIT) have developed a machine learning approach that could speed up vehicle routing algorithms, which tend to slow down when applied to large urban datasets.
The new "learning-to-delegate" approach could increase the speed of the strongest algorithmic solvers by determining the most useful subproblems to solve, rather than having the algorithm solve all subproblems.
The researchers created a neural network that automatically identifies those subproblems that, when solved, result in the greatest gain in solution quality. The researchers said the approach could be used with a variety of solvers and resource allocation problems.
Said MIT's Cathy Wu, "We may unlock new applications that now will be possible because the cost of solving the problem is 10 to 100 times less."
From MIT News
View Full Article
Abstracts Copyright © 2021 SmithBucklin, Washington, DC, USA
No entries found