PHEW : Constructing Sparse Networks that Learn Fast and Generalize Well without Training Data.
Shreyas Malakarjun PatilConstantine DovrolisPublished in: ICML (2021)
Keyphrases
- training data
- training set
- learning algorithm
- supervised learning
- high dimensional
- data sets
- test data
- prior knowledge
- domain knowledge
- training instances
- sparse data
- complex networks
- avoid overfitting
- sparse matrix
- cellular networks
- random projections
- training dataset
- noisy data
- complex systems
- class labels
- sparse representation
- test set
- generative model
- small number