Leveraging Sub-Optimal Data for Human-in-the-Loop Reinforcement Learning.
Calarina MuslimaniMatthew E. TaylorPublished in: CoRR (2024)
Keyphrases
- data sets
- reinforcement learning
- high quality
- prior knowledge
- training data
- database
- data collection
- data processing
- complex data
- raw data
- data sources
- xml documents
- data analysis
- data structure
- original data
- data distribution
- experimental data
- statistical analysis
- computer systems
- input data
- end users
- knowledge discovery
- learning algorithm
- machine learning
- decision trees
- spatial data
- dynamic programming
- wireless sensor networks
- network structure
- statistical methods
- human experts
- neural network
- optimal solution
- data objects
- databases