Deriving Rewards for Reinforcement Learning from Symbolic Behaviour Descriptions of Bipedal Walking.
Daniel HarnackChristoph LüthLukas GrossShivesh KumarFrank KirchnerPublished in: CDC (2023)
Keyphrases
- reinforcement learning
- reward function
- symbolic description
- markov decision processes
- high level
- reinforcement learning algorithms
- machine learning
- state space
- function approximation
- symbolic descriptions
- learning algorithm
- symbolic representation
- optimal policy
- temporal difference
- multi agent
- multiple agents
- model free
- partially observable
- transfer learning
- function approximators
- average reward
- semantic description
- natural language descriptions
- supervised learning
- reward shaping
- learning tasks
- action selection
- symbolic data