arXivDaily arXiv每日学术速递 周一至周五更新
arXiv周末暂无论文更新,休息一下吧,周末愉快~~
arXiv 2210.07686cs.LGcs.AI
收藏

基于知识蒸馏的车辆路径问题广义模型学习

Learning Generalizable Models for Vehicle Routing Problems via Knowledge Distillation

Sun Yat-sen University(中山大学);National University of Singapore(新加坡国立大学);Singapore Institute of Manufacturing Technology, A*STAR(新加坡科技研究局新加坡制造技术研究院);University of Melbourne(墨尔本大学)

Jieyi Bi, Yining Ma, Jiahai Wang, Zhiguang Cao, Jinbiao Chen, Yuan Sun, Yeow Meng Chee

摘要:

最近用于车辆路径问题的神经方法总是在相同的实例分布上训练和测试深度模型(即,制服)。为了解决随之而来的交叉分布泛化问题,我们将知识蒸馏引入这一领域,并提出了一种自适应多分布知识蒸馏(AMDKD)方案,用于学习更可泛化的深度模型。特别是,我们的AMDKD利用了来自多位教师的各种知识,这些教师都接受过样本分布的培训,从而产生了一个轻量级但又通才的学生模型。同时,我们为AMDKD配备了自适应策略,允许学生专注于困难的分布,以便更有效地吸收难以掌握的知识。大量的实验结果表明,与基线神经方法相比,我们的AMDKD能够在看不见的分布内和分布外实例上实现有竞争力的结果,这些实例是随机合成的或从基准数据集(即,TSPLIB和CVRPLIB)。值得注意的是,我们的AMDKD是通用的,并且消耗较少的计算资源用于推理。

英文摘要:

Recent neural methods for vehicle routing problems always train and test the deep models on the same instance distribution (i.e., uniform). To tackle the consequent cross-distribution generalization concerns, we bring the knowledge distillation to this field and propose an Adaptive Multi-Distribution Knowledge Distillation (AMDKD) scheme for learning more generalizable deep models. Particularly, our AMDKD leverages various knowledge from multiple teachers trained on exemplar distributions to yield a light-weight yet generalist student model. Meanwhile, we equip AMDKD with an adaptive strategy that allows the student to concentrate on difficult distributions, so as to absorb hard-to-master knowledge more effectively. Extensive experimental results show that, compared with the baseline neural methods, our AMDKD is able to achieve competitive results on both unseen in-distribution and out-of-distribution instances, which are either randomly synthesized or adopted from benchmark datasets (i.e., TSPLIB and CVRPLIB). Notably, our AMDKD is generic, and consumes less computational resources for inference.

机构、总结、中文摘要均由AI分析而来,可能存在错误,请以论文正文为准。

↑