Prompt Learning for Generalized Vehicle Routing

Authors: Fei Liu, Xi Lin, Weiduo Liao, Zhenkun Wang, Qingfu Zhang, Xialiang Tong, Mingxuan Yuan

IJCAI 2024 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments show that the proposed prompt learning approach facilitates the fast adaptation of pre-trained routing models. It also outperforms existing generalized models on both in-distribution prediction and zero-shot generalization to a diverse set of new tasks.
Researcher Affiliation Collaboration Fei Liu1, Xi Lin1, Weiduo Liao1, Zhenkun Wang2, , Qingfu Zhang1, , Xialiang Tong3, and Mingxuan Yuan3 1City University of Hong Kong 2Southern University of Science and Technology 3Huawei Noah s Ark Lab
Pseudocode No The paper does not contain structured pseudocode or algorithm blocks.
Open Source Code Yes Our code implementation is available online at https://github.com/Fei Liu36/Prompt VRP.
Open Datasets Yes The detailed instance generation is introduced in the Appendix, which is the same as that used by Zhou [2023]. We adopt the distribution proposed by Bi [2022], which consists of a total of 12 different distributions, including cluster (CL), expansion (EA), explosion (EO), implosion (IM), grid (GR), and mixed (MX). More experiments on real-world instances are conducted on five test suites: set A, B, P, X [Uchoa et al., 2017], and XML [Queiroga et al., 2021] from CVRPLIB 1. 1http://vrp.atd-lab.inf.puc-rio.br/
Dataset Splits No The paper describes the instance generation and uses training distributions and test distributions, but it does not explicitly specify a separate 'validation' dataset split with percentages or counts.
Hardware Specification Yes It takes only about 2.5 days on a single RTX 2080Ti with 11 GB GPU memory.
Software Dependencies No The paper mentions algorithms and pre-trained models used (e.g., REINFORCE, Attention model, POMO), but does not specify software packages or libraries with version numbers (e.g., Python, PyTorch, TensorFlow, or specific library versions).
Experiment Setup Yes The prompt size is set to be M = 16, and the number of prompted tokens for each layer is D = 5. As there are L = 6 MHA layers in the encoder and the embedding size for each token is E = 128, the lengths of the key and prompt vectors are 6 128 = 768 and 5 6 128 = 3840, respectively. The prompts are trained with a batch size of B = 64. The maximum number of epochs is 10, 000 and there are 1, 000 instances for each epoch. The learning rate decays from 1e 3 to 1e 5.