User: Guest  Login
Document type:
Zeitschriftenaufsatz
Author(s):
Pirnay, Jonathan; Grimm, Dominik G.
Title:
Self-Improvement for Neural Combinatorial Optimization: Sample Without Replacement, but Improvement
Abstract:
Current methods for end-to-end constructive neural combinatorial optimization usually train a policy using behavior cloning from expert solutions or policy gradient methods from reinforcement learning. While behavior cloning is straightforward, it requires expensive expert solutions, and policy gradient methods are often computationally demanding and complex to fine-tune. In this work, we bridge the two and simplify the training process by sampling multiple solutions for random instances using t...     »
Keywords:
Machine Learning, Self-Improvement Learning
Journal title:
Transactions on Machine Learning Research
Year:
2024
Year / month:
2024-06
Reviewed:
ja
WWW:
https://openreview.net/forum?id=agT8ojoH0X
Print-ISSN:
2835-8856
Notes:
Featured Certification
 BibTeX