Learning from Human Feedback: Challenges for Real-World Reinforcement Learning in NLP.
Julia KreutzerStefan RiezlerCarolin LawrencePublished in: CoRR (2020)
Keyphrases
- reinforcement learning
- real world
- learning algorithm
- learning process
- multi agent
- motor skills
- learning capabilities
- data sets
- online learning
- language acquisition
- learning problems
- natural language processing
- supervised learning
- knowledge acquisition
- markov decision processes
- learning tasks
- knowledge representation
- human experts
- active learning
- prior knowledge
- temporal difference learning
- reinforcement learning methods
- natural language
- natural language learning
- creative problem solving