Transfer learning for atomistic simulations using GNNs and kernel mean embeddings.
John Isak Texas FalkLuigi BonatiPietro NovelliMichele ParrinelloMassimiliano PontilPublished in: CoRR (2023)
Keyphrases
- transfer learning
- latent space
- learning tasks
- labeled data
- machine learning
- knowledge transfer
- reinforcement learning
- semi supervised learning
- cross domain
- manifold alignment
- kernel function
- multi task
- collaborative filtering
- active learning
- text categorization
- kernel methods
- transfer knowledge
- support vector
- text classification
- feature space
- machine learning algorithms
- vector space
- euclidean space
- semi supervised
- gaussian processes
- target domain
- unlabeled data
- data sets
- learning algorithm
- databases
- low dimensional
- domain specific
- dimensionality reduction
- feature selection
- domain adaptation
- kernel matrix
- neural network