t-DGR: A Trajectory-Based Deep Generative Replay Method for Continual Learning in Decision Making.
William YueBo LiuPeter StonePublished in: CoRR (2024)
Keyphrases
- detection method
- decision making
- learning algorithm
- high accuracy
- pairwise
- learning scheme
- unsupervised learning
- classification accuracy
- computational cost
- preprocessing
- learning process
- significant improvement
- cost function
- online learning
- similarity measure
- neural network
- active learning
- dynamic programming
- objective function
- supervised learning
- reinforcement learning
- markov random field
- training data