Cross-Problem Learning for Solving Vehicle Routing Problems

Authors: Zhuoyi Lin, Yaoxin Wu, Bangjian Zhou, Zhiguang Cao, Wen Song, Yingqian Zhang, Senthilnath Jayavelu

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

Reproducibility Variable Result LLM Response
Research Type Experimental Extensive experiments on typical VRPs substantiate that 1) the full fine-tuning achieves significantly better performance than the one trained from scratch, and 2) the adapter-based fine-tuning also delivers comparable performance while being notably parameterefficient. Furthermore, we empirically demonstrate the favorable effect of our method in terms of cross-distribution application and versatility.
Researcher Affiliation Collaboration Zhuoyi Lin1 , Yaoxin Wu2 , Bangjian Zhou3 , Zhiguang Cao4 , Wen Song5 , Yingqian Zhang2 and Senthilnath Jayavelu1 1Institute for Infocomm Research, Agency for Science, Technology and Research (A*STAR) 2Eindhoven University of Technology 3Unidt 4Singapore Management University 5Shandong University
Pseudocode No The paper describes its methods narratively and with figures, but does not include structured pseudocode or algorithm blocks.
Open Source Code Yes The code is available: https://github.com/Zhuoyi-Lin/ Cross problem learning.
Open Datasets No Following the existing works, we generate the coordinates of nodes by uniformly sampling from the unit square [0, 1] [0, 1]. For OP, we set prizes all equal to 1. For PCTSP, we randomly sample prizes and penalties from Uniform distributions, following [Kool et al., 2018].
Dataset Splits Yes For each epoch, the training instances are generated on the fly, and we use 10,000 instances for the validation.
Hardware Specification Yes The server with Intel Core i9-10940X CPU and NVIDIA Ge Force RTX 2080Ti is used in our experiments.
Software Dependencies No The paper mentions software like 'Transformer based AM' and 'POMO' but does not specify version numbers for any key software components (e.g., Python, PyTorch, CUDA).
Experiment Setup Yes Specifically, we train the model with 2500 batches in each epoch and use 512 instances in each batch (except for the problem with n = 100, where we use 256 instances). For each epoch, the training instances are generated on the fly, and we use 10,000 instances for the validation. Then the paired t-test (α = 5%) is conducted after each epoch to replace the baseline model, according to the improvement significance. For all problems and sizes, we train 100 epochs for the full convergence of the training curves. The learning rate is set as 10 4 in all our methods...