notesum.ai

Published at October 21

Offline reinforcement learning for job-shop scheduling problems

cs.LG

cs.AI

Released Date: October 21, 2024

Authors: Imanol Echeverria¹, Maialen Murua¹, Roberto Santana²

Aff.: ¹TECNALIA, Basque Research and Technology Alliance (BRTA); ²Computer Science and Artificial Intelligence Department, University of the Basque Country

Arxiv: https://arxiv.org/abs/2410.15714v1

Method	Problem	Learning method	Algorithm	Type	Year
[12]	JSSP	RL	PPO	Constructive	2020
[13]	JSSP	RL	REINFORCE	Constructive	2021
[14]	JSSP	RL	Policy gradient	Constructive	2022
[15]	JSSP	RL	Policy gradient	Constructive	2023
[16]	JSSP	RL	REINFORCE	Improvement	2024
[17]	JSSP, FJSSP	RL	REINFORCE	Constructive	2024
[18]	FJSSP	RL	PPO	Constructive	2022
[19]	FJSSP	RL	PPO	Constructive	2023
[20]	FJSSP	RL	PPO	Constructive	2023
[21]	FJSSP	RL	PPO	Constructive	2024
[22]	FJSSP	RL	PPO	Constructive	2024
[23]	JSSP	BC	-	Constructive	2018
[24]	FSS	BC	-	Constructive	2022
[25]	FJSSP	BC	-	Constructive	2024
[26]	JSSP	Self-supervised	-	Constructive	2024
Ours	JSSP, FJSSP	Offline RL	TD3 + BC	Constructive	2024