Abstract
Traveling Salesman Problem (TSP) and similar combinatorial search and optimization problems have many real-world applications in logistics, transportation, manufacturing, IC design, and other industries. Large-scale TSP tasks have always been challenging to solve fast. During the training phase of the model, when the number of city nodes exceeds 200, the training process will be terminated due to insufficient memory. This paper achieves reducing memory usage by simplifying the network model. However, the prediction accuracy is lowered after the network model is simplified. In this paper, heuristic search methods such as greedy search, beam search and 2-opt search are used to improve the prediction accuracy. Our main contributions are: increase the number of city nodes that can be solved from 100 to 1000; compensate for the loss of accuracy with various search techniques; use various search techniques in combinatorial search and optimization domain. The novelty of our paper is: the model structure of the Transformer is simplified, and various heuristic search techniques are used to compensate for the accuracy of the solution. In the inference stage, although the search time required by greedy search, beam search, and 2-opt search is quite different, all of them can improve the model’s prediction accuracy to varying degrees. Extensive experiments demonstrate that using various heuristic search techniques can greatly improve the prediction accuracy of the model.
Access this chapter
Tax calculation will be finalised at checkout
Purchases are for personal use only
References
Arora, S.: The approximability of NP-hard problems. In: Proceedings of the Thirtieth Annual ACM Symposium on Theory of Computing, pp. 337–348 (1998)
Bahdanau, D., Cho, K.H., Bengio, Y.: Neural machine translation by jointly learning to align and translate. In: 3rd International Conference on Learning Representations, ICLR 2015 (2015)
Bello, I., Pham, H., Le, Q.V., Norouzi, M., Bengio, S.: Neural combinatorial optimization with reinforcement learning. arXiv preprint arXiv:1611.09940 (2016)
Boese, K.D.: Cost versus distance in the traveling salesman problem. Citeseer (1995)
Bresson, X., Laurent, T.: The transformer network for the traveling salesman problem. arXiv preprint arXiv:2103.03012 (2021)
Cook, W., Lovász, L., Seymour, P.D., et al.: Combinatorial optimization: papers from the DIMACS Special Year, vol. 20. American Mathematical Soc. (1995)
Deudon, M., Cournut, P., Lacoste, A., Adulyasak, Y., Rousseau, L.-M.: Learning heuristics for the TSP by policy gradient. In: van Hoeve, W.-J. (ed.) CPAIOR 2018. LNCS, vol. 10848, pp. 170–181. Springer, Cham (2018). https://doi.org/10.1007/978-3-319-93031-2_12
Gasse, M., Chételat, D., Ferroni, N., Charlin, L., Lodi, A.: Exact combinatorial optimization with graph convolutional neural networks. In: Advances in Neural Information Processing Systems, vol. 32 (2019)
Google, I.: Google optimization tools (or-tools) (2018). https://github.com/google/or-tools
Gutin, G., Punnen, A.P.: The Traveling Salesman Problem and Its Variations, vol. 12. Springer, New York (2006). https://doi.org/10.1007/b101971
Helsgaun, K.: An extension of the Lin-Kernighan-Helsgaun TSP solver for constrained traveling salesman and vehicle routing problems. Technical report (2017)
Hochba, D.S.: Approximation algorithms for np-hard problems. ACM SIGACT News 28(2), 40–52 (1997)
Hopfield, J.J., Tank, D.W.: Neural computation of decisions in optimization problems. Biol. Cybern. 52(3), 141–152 (1985)
Johnson, D.: Local search and the traveling salesman problem. In: Automata Languages and Programming. LNCS, pp. 443–460. Springer, Berlin (1990)
Joshi, C.K., Cappart, Q., Rousseau, L.M., Laurent, T., Bresson, X.: Learning TSP requires rethinking generalization. arXiv preprint arXiv:2006.07054 (2020)
Jünger, M., Reinelt, G., Rinaldi, G.: The traveling salesman problem. In: Handbooks in Operations Research and Management Science, vol. 7, pp. 225–330 (1995)
Khalil, E., Dai, H., Zhang, Y., Dilkina, B., Song, L.: Learning combinatorial optimization algorithms over graphs. In: Advances in Neural Information Processing Systems, pp. 6348–6358 (2017)
Kool, W., van Hoof, H., Welling, M.: Attention, learn to solve routing problems! In: International Conference on Learning Representations (2018)
Li, W., Ding, Y., Yang, Y., Sherratt, R.S., Park, J.H., Wang, J.: Parameterized algorithms of fundamental np-hard problems: a survey. HCIS 10(1), 1–24 (2020)
Lin, S., Kernighan, B.W.: An effective heuristic algorithm for the traveling-salesman problem. Oper. Res. 21(2), 498–516 (1973)
Ma, Q., Ge, S., He, D., Thaker, D., Drori, I.: Combinatorial optimization by graph pointer networks and hierarchical reinforcement learning. arXiv preprint arXiv:1911.04936 (2019)
Nazari, M., Oroojlooy, A., Snyder, L., Takác, M.: Reinforcement learning for solving the vehicle routing problem. In: Advances in Neural Information Processing Systems, pp. 9839–9849 (2018)
Gurobi Optimization: Gurobi optimizer reference manual (2018). http://www.gurobi.com
Papadimitriou, C.H., Steiglitz, K.: Combinatorial optimization: algorithms and complexity. Courier Corporation (1998)
Peng, B., Wang, J., Zhang, Z.: A deep reinforcement learning algorithm using dynamic attention model for vehicle routing problems. In: Li, K., Li, W., Wang, H., Liu, Y. (eds.) ISICA 2019. CCIS, vol. 1205, pp. 636–650. Springer, Singapore (2020). https://doi.org/10.1007/978-981-15-5577-0_51
Sutton, R.S., Barto, A.G.: Reinforcement Learning: An Introduction. MIT Press, Cambridge (2018)
Chvatal, V., Applegate, D.L., Bixby, R.E., Cook, W.J.: Concorde TSP solver (2006). www.math.uwaterloo.ca/tsp/concorde
Vaswani, A., et al.: Attention is all you need. In: Advances in Neural Information Processing Systems, pp. 5998–6008 (2017)
Vinyals, O., Fortunato, M., Jaitly, N.: Pointer networks. Comput. Sci. 28 (2015)
Welling, M., Kipf, T.N.: Semi-supervised classification with graph convolutional networks. In: International Conference on Learning Representations (ICLR 2017) (2016)
Williams, R.J.: Simple statistical gradient-following algorithms for connectionist reinforcement learning. Mach. Learn. 8(3), 229–256 (1992)
Woeginger, G.J.: Exact algorithms for NP-hard problems: a survey. In: Jünger, M., Reinelt, G., Rinaldi, G. (eds.) Combinatorial Optimization — Eureka, You Shrink! LNCS, vol. 2570, pp. 185–207. Springer, Heidelberg (2003). https://doi.org/10.1007/3-540-36478-1_17
Xing, Z., Tu, S., Xu, L.: Solve traveling salesman problem by Monte Carlo tree search and deep neural network. arXiv preprint arXiv:2005.06879 (2020)
Yang, H.: Extended attention mechanism for tsp problem. In: 2021 International Joint Conference on Neural Networks (IJCNN), pp. 1–8. IEEE (2021)
Author information
Authors and Affiliations
Corresponding author
Editor information
Editors and Affiliations
Rights and permissions
Copyright information
© 2023 The Author(s), under exclusive license to Springer Nature Singapore Pte Ltd.
About this paper
Cite this paper
Yang, H., Gu, M. (2023). Learning TSP Combinatorial Search and Optimization with Heuristic Search. In: Tanveer, M., Agarwal, S., Ozawa, S., Ekbal, A., Jatowt, A. (eds) Neural Information Processing. ICONIP 2022. Communications in Computer and Information Science, vol 1791. Springer, Singapore. https://doi.org/10.1007/978-981-99-1639-9_34
Download citation
DOI: https://doi.org/10.1007/978-981-99-1639-9_34
Published:
Publisher Name: Springer, Singapore
Print ISBN: 978-981-99-1638-2
Online ISBN: 978-981-99-1639-9
eBook Packages: Computer ScienceComputer Science (R0)