APSL: Action-positive separation learning for unsupervised temporal action localization.
Yuanyuan LiuNing ZhouFayong ZhangWenbin WangYu WangKejun LiuZiyuan LiuPublished in: Inf. Sci. (2023)
Keyphrases
- action models
- action selection
- unsupervised learning
- supervised learning
- learning from experience
- neural network
- active learning
- learning process
- training data
- spatio temporal
- reinforcement learning
- online learning
- knowledge acquisition
- positive and negative
- learning algorithm
- learning problems
- data sets
- partial observations
- machine learning
- action sequences
- positive examples
- learning tasks
- spatial and temporal
- domain independent
- learning environment
- prior knowledge