MERGE: Meta Reinforcement Learning for Tunable RL Agents at the Edge.
Sharda TripathiCarla-Fabiana ChiasseriniPublished in: GLOBECOM (2023)
Keyphrases
- reinforcement learning
- multi agent
- learning agents
- action selection
- single agent
- learning agent
- multi agent environments
- function approximation
- multi agent reinforcement learning
- reinforcement learning agents
- multi agent systems
- multiagent systems
- intelligent agents
- reinforcement learning algorithms
- robocup soccer
- markov decision processes
- state space
- agent learns
- temporal difference
- model free
- mobile agents
- multiple agents
- edge information
- autonomous agents
- partial observability
- software agents
- edge detection
- cooperative
- rl algorithms
- multiagent learning
- optimal policy
- multiagent reinforcement learning
- learning capabilities
- learning algorithm
- agent receives
- machine learning
- action space
- meta reasoning
- autonomous learning
- dynamic environments
- learned knowledge
- decision making
- dynamic programming
- direct policy search
- complex domains
- partially observable domains
- dynamical systems
- learning problems
- partially observable
- stochastic games
- actor critic
- state abstraction
- decision problems
- evaluation function
- learning process
- learning classifier systems
- reward shaping
- temporal difference learning
- control policy