Learning to Iteratively Solve Routing Problems with Dual-Aspect Collaborative Transformer
arXiv:2110.02544
Abstract
Recently, Transformer has become a prevailing deep architecture for solving vehicle routing problems (VRPs). However, it is less effective in learning improvement models for VRP because its positional encoding (PE) method is not suitable in representing VRP solutions. This paper presents a novel Dual-Aspect Collaborative Transformer (DACT) to learn embeddings for the node and positional features separately, instead of fusing them together as done in existing ones, so as to avoid potential noises and incompatible correlations. Moreover, the positional features are embedded through a novel cyclic positional encoding (CPE) method to allow Transformer to effectively capture the circularity and symmetry of VRP solutions (i.e., cyclic sequences). We train DACT using Proximal Policy Optimization and design a curriculum learning strategy for better sample efficiency. We apply DACT to solve the traveling salesman problem (TSP) and capacitated vehicle routing problem (CVRP). Results show that our DACT outperforms existing Transformer based improvement models, and exhibits much better generalization performance across different problem sizes on synthetic and benchmark instances, respectively.
arXiv commentAccepted at NeurIPS 2021. Changes in v3: updated CVRP performance due to more accurate feasibility masks (reduced training time); improved comparison analysis in Table 1 and Section 5; fixed minor typos/inaccuracies in the CPE equation, Figure 6, and other sentences
References in corpus (10)
- Distilling the Knowledge in a Neural Network
- Learning Combinatorial Optimization Algorithms over Graphs
- Learning to Dispatch for Job Shop Scheduling via Deep Reinforcement Learning
- Deep Reinforcement Learning for Solving the Heterogeneous Capacitated Vehicle Routing Problem
- An Efficient Graph Convolutional Network Technique for the Travelling Salesman Problem
- Attention, Learn to Solve Routing Problems!
- Heterogeneous Attentions for Solving Pickup and Delivery Problem via Deep Reinforcement Learning
- Implementation Matters in Deep Policy Gradients: A Case Study on PPO and TRPO
- Neural Large Neighborhood Search for the Capacitated Vehicle Routing Problem
- Multi-Decoder Attention Model with Embedding Glimpse for Solving Vehicle Routing Problems