Pre-Training Acquisition Functions by Deep Reinforcement Learning for Fixed Budget Active Learning.
Yusuke TaguchiHideitsu HinoKeisuke KameyamaPublished in: Neural Process. Lett. (2021)
Keyphrases
- active learning
- reinforcement learning
- training examples
- supervised learning
- batch mode
- training set
- learning algorithm
- annotation effort
- learning process
- transfer learning
- test set
- random sampling
- deep architectures
- function approximation
- state space
- multi agent
- training process
- experimental design
- training phase
- learning strategies
- selective sampling
- sample selection
- active exploration
- pool based active learning
- imbalanced data classification
- learning tasks
- data sets
- training algorithm
- reinforcement learning algorithms
- query by committee