A continual learning framework to train robust image recognition models by adversarial training and knowledge distillation.
Ting-Chun ChouYu-Cheng KuoJhih-Yuan HuangWei-Po LeePublished in: Connect. Sci. (2024)
Keyphrases
- image recognition
- learning systems
- prior knowledge
- structured prediction
- probabilistic model
- deep architectures
- knowledge acquisition
- knowledge transfer
- learning scheme
- design principles
- learning process
- learning algorithm
- bayesian framework
- conceptual framework
- dynamic bayesian networks
- previously learned
- background knowledge
- multi agent
- supervised learning
- learning tasks
- constructivist learning environments
- face recognition
- accurate models
- active learning
- domain knowledge
- image classification
- learning models
- concept maps
- object detectors
- expert systems
- learning mechanism
- qualitative models
- learned models
- online learning
- neural network
- computer based training
- learning frameworks
- knowledge base
- semi supervised learning