Adaptive Agents in Minecraft: A Hybrid Paradigm for Combining Domain Knowledge with Reinforcement Learning.
Priyam ParasharBradley ShenemanAshok K. GoelPublished in: AAMAS Workshops (Visionary Papers) (2017)
Keyphrases
- domain knowledge
- reinforcement learning
- multi agent
- learning capabilities
- multi agent systems
- learning agents
- adaptive behavior
- learned knowledge
- multiagent systems
- action selection
- multiple agents
- multi agent reinforcement learning
- agent architecture
- autonomous agents
- agent receives
- model free
- intelligent agents
- knowledge sources
- function approximation
- robocup soccer
- coalition formation
- learning algorithm
- learning agent
- background knowledge
- reinforcement learning agents
- mobile agents
- partial observability
- single agent
- cooperative
- reasoning process
- neural network
- decision making
- software agents
- dynamic environments
- electronic commerce
- agent systems
- artificial agents
- machine learning
- human users
- agent technology
- incomplete information