Overfreezing Meets Overparameterization: A Double Descent Perspective on Transfer Learning of Deep Neural Networks.
Yehuda DarLorenzo LuziRichard G. BaraniukPublished in: CoRR (2022)
Keyphrases
- transfer learning
- neural network
- learning tasks
- algorithmic information theory
- knowledge transfer
- cross domain
- reinforcement learning
- labeled data
- pattern recognition
- transfer knowledge
- fuzzy logic
- structure learning
- semi supervised learning
- active learning
- multi task learning
- machine learning algorithms
- manifold alignment
- collaborative filtering
- domain adaptation
- learning algorithm
- machine learning
- text mining
- text classification
- text categorization
- target domain
- multi task
- unlabeled data
- transferring knowledge
- data sets
- face images
- bayesian networks
- multitask learning
- previously learned
- supervised learning
- knn