Unsupervised Learning for Solving the Travelling Salesman Problem

Authors: Yimeng Min, Yiwei Bai, Carla P. Gomes

NeurIPS 2023 | Conference PDF | Archive PDF | Plain Text | LLM Run Details

Reproducibility Variable Result LLM Response
Research Type Experimental Experimental results show that UTSP outperforms the existing data-driven TSP heuristics.
Researcher Affiliation Academia Yimeng Min Dept. of Computer Science Cornell University Ithaca, NY, USA min@cs.cornell.edu Yiwei Bai Dept. of Computer Science Cornell University Ithaca, NY, USA bywbilly@gmail.com Carla P. Gomes Dept. of Computer Science Cornell University Ithaca, NY, USA gomes@cs.cornell.edu
Pseudocode No The paper describes its methods through prose and mathematical equations but does not include any explicit pseudocode or algorithm blocks.
Open Source Code No The paper does not provide any statement or link indicating the availability of open-source code for the described methodology.
Open Datasets Yes Our dataset contains 2,000 samples for training and 1,000 samples for validation. We use the same test dataset in Fu et al. [2021].
Dataset Splits Yes Our dataset contains 2,000 samples for training and 1,000 samples for validation.
Hardware Specification Yes All models are trained using Nvidia V100 GPU.
Software Dependencies No The paper mentions using Adam optimizer but does not specify any software names with version numbers for reproducibility (e.g., Python, PyTorch, TensorFlow versions).
Experiment Setup Yes All the search-related parameters are listed in Table 4. M is the size of the candidate set of each city. K is the maximal number of edges we can remove in one action, and for each round of local search, we randomly select one number from the listed interval. T is the total number of actions we will try to expand one node.