UvA-DARE ( Digital Academic Repository ) Attention , learn to solve routing problems ! | Semantic Scholar (2024)

This article proposes a deep reinforcement learning framework to learn the improvement heuristics for routing problems, and designs a self-attention-based deep architecture as the policy network to guide the selection of the next solution.

This work proposes a novel large neighborhood search (LNS) framework for vehicle routing that integrates learned heuristics for generating new solutions and shows for the CVRP and the SDVRP that this approach surpasses the performance of existing machine learning approaches and comes close to theperformance of state-of-the-art optimization approaches.

This work proposes a policy gradient algorithm to learn a stochastic policy that selects 2-opt operations given a current solution and introduces a policy neural network that leverages a pointing attention mechanism, which can be easily extended to more general k-opt moves.

A Neural Improvement model for graph-based combinatorial problems that, given an instance and a candidate solution, encodes the problem information by means of edge features and can be a milestone in the development of trajectory-based optimization algorithms.

This paper introduces a novel deep learning approach for approximately solving the most famous NP-hard problem in recent history, the Travelling Salesman Problem and focuses on the 2D Euclidean TSP and uses Graph Convolutional Neural Networks and beam search to predict a valid TSP tour given an input graph with up to 100 nodes.

A novel framework is proposed (called Alpha-T) based on AlphaZero, which does not require expert experience or label data but is trained through self-play and divides the learning into two stages: in the first stage it employs graph attention network (GAT) and GRU to learn node representations and memory history trajectories, and in the second stage it uses Monte Carlo tree search (MCTS) and deep RL to search the solution space and train the model.

This paper presents a reinforcement learning formulation for the improvement heuristic, where the policy guides selection of the next solution, and proposes a deep architecture as the policy network based on self-attention.

This work proposes a policy gradient algorithm to learn a stochastic policy that selects 2-opt operations given a current solution and introduces a policy neural network that leverages a pointing attention mechanism, which unlike previous works, can be easily extended to more general k-opt moves.

This work proposes to adapt the recent successful Transformer architecture originally developed for natural language processing to the combinatorial TSP, and reports improved performances over recent learned heuristics.

...

...

The neural combinatorial optimization framework is extended to solve the traveling salesman problem (TSP) and the performance of the proposed framework alone is generally as good as high performance heuristics (OR-Tools).

A neural network solution that treats the salesmen, cities and depot as three different sets of varying cardinalities is designed and shown to outperform all the meta-heuristics of the leading solver in the field.

This paper proposes a unique combination of reinforcement learning and graph embedding that behaves like a meta-algorithm that incrementally constructs a solution, and the action is determined by the output of agraph embedding network capturing the current state of the solution.

This work presents an end-to-end framework for solving the Vehicle Routing Problem (VRP) using reinforcement learning, and demonstrates how this approach can handle problems with split delivery and explore the effect of such deliveries on the solution quality.

This revised note is interested in studying another aspect of hardness, related to the ability to learn how to solve a problem by simply observing a collection of previously solved instances, which is illustrated on the Quadratic Assignment Problem.

A framework to tackle combinatorial optimization problems using neural networks and reinforcement learning, and Neural Combinatorial Optimization achieves close to optimal results on 2D Euclidean graphs with up to 100 nodes.

A branch-and-cut algorithm for finding an optimal OP solution, based on several families of valid inequalities, is described and proved to be able to solve to optimality large-scale instances involving up to 500 nodes, within acceptable computing time.

This paper considers the problem of optimizing image captioning systems using reinforcement learning, and shows that by carefully optimizing systems using the test metrics of the MSCOCO task, significant gains in performance can be realized.

The implementation of an extension of the Lin-Kernighan-Helsgaun TSP solver, which is able to solve a variety of well-known problems, including the sequential ordering problem, the traveling repairman problem, variants of the multiple travel-ing salesman problem, as well as vehicle routing problems with capacity, time windows, pickup-and-delivery and distance constraints.

...

...