Deterministic equivalent and error universality of deep random features learning.
Dominik SchröderHugo CuiDaniil DmitrievBruno LoureiroPublished in: CoRR (2023)
Keyphrases
- learning algorithm
- learning process
- reinforcement learning
- learning systems
- unsupervised feature learning
- deep learning
- feature extraction
- prior knowledge
- image features
- feature set
- learning tasks
- feature space
- lower bound
- feature vectors
- supervised learning
- error rate
- decision trees
- false positives
- feature selection
- neural network