A Semi-Markov Decision Model With Inverse Reinforcement Learning for Recognizing the Destination of a Maneuvering Agent in Real Time Strategy Games.
Yunxiu ZengKai XuLong QinQuanjun YinPublished in: IEEE Access (2020)
Keyphrases
- real time strategy games
- decision model
- semi markov
- inverse reinforcement learning
- multi agent
- decision problems
- decision making
- case based planning
- decision makers
- reward function
- preference elicitation
- conditional random fields
- reinforcement learning
- imperfect information
- decision theory
- multi agent systems
- cooperative
- partially observable
- influence diagrams
- multiagent systems
- objective function