Search Results for author: André Hottung

Found 7 papers, 4 papers with code

PolyNet: Learning Diverse Solution Strategies for Neural Combinatorial Optimization

no code implementations21 Feb 2024 André Hottung, Mridul Mahajan, Kevin Tierney

Reinforcement learning-based methods for constructing solutions to combinatorial optimization problems are rapidly approaching the performance of human-designed algorithms.

Combinatorial Optimization

Simulation-guided Beam Search for Neural Combinatorial Optimization

1 code implementation13 Jul 2022 Jinho Choo, Yeong-Dae Kwon, Jihoon Kim, Jeongwoo Jae, André Hottung, Kevin Tierney, Youngjune Gwon

Neural approaches for combinatorial optimization (CO) equip a learning mechanism to discover powerful heuristics for solving complex real-world problems.

Combinatorial Optimization

Efficient Active Search for Combinatorial Optimization Problems

2 code implementations ICLR 2022 André Hottung, Yeong-Dae Kwon, Kevin Tierney

While active search is simple to implement, it is not competitive with state-of-the-art methods because adjusting all model weights for each test instance is very time and memory intensive.

BIG-bench Machine Learning Combinatorial Optimization +2

Neural Large Neighborhood Search for the Capacitated Vehicle Routing Problem

1 code implementation21 Nov 2019 André Hottung, Kevin Tierney

Furthermore, we show for the CVRP and the SDVRP that our approach surpasses the performance of existing machine learning approaches and comes close to the performance of state-of-the-art optimization approaches.

BIG-bench Machine Learning

Deep Learning Assisted Heuristic Tree Search for the Container Pre-marshalling Problem

no code implementations28 Sep 2017 André Hottung, Shunji Tanaka, Kevin Tierney

The container pre-marshalling problem (CPMP) is concerned with the re-ordering of containers in container terminals during off-peak times so that containers can be quickly retrieved when the port is busy.

Cannot find the paper you are looking for? You can Submit a new open access paper.