The StarCraft Multi-Agent Exploration Challenges: Learning Multi-Stage Tasks and Environmental Factors Without Precise Reward Functions.
Mingyu KimJihwan OhYongsik LeeJoonkee KimSeonghwan KimSong ChongSeyoung YunPublished in: IEEE Access (2023)
Keyphrases
- multistage
- reinforcement learning
- multi agent
- environmental factors
- active learning
- inverse reinforcement learning
- learning agents
- multiple agents
- optimal policy
- stochastic programming
- learning algorithm
- machine learning
- graphical models
- lot sizing
- dynamic programming
- reward function
- prior knowledge
- single stage
- information technology
- case study
- assembly systems