Intelligence arms race: delayed reward increases complexity of agent strategies.
Hirotaka OsawaPublished in: AAMAS (2014)
Keyphrases
- expected reward
- multi agent
- multi agent systems
- multi armed bandit problems
- autonomous agents
- reinforcement learning
- computational complexity
- agent architecture
- artificial intelligence
- software agents
- agent receives
- intelligent software agents
- trading agents
- reward function
- space complexity
- agent technology
- markov decision processes
- intelligent agents
- dynamic environments
- decision making
- resource allocation
- mobile agents
- long run
- machine intelligence
- action selection
- agent systems
- learning agents
- information processing
- intelligent systems
- supply chain