Asymptotics of feature learning in two-layer networks after one gradient-step.
Hugo CuiLuca PesceYatin DandiFlorent KrzakalaYue M. LuLenka ZdeborováBruno LoureiroPublished in: CoRR (2024)
Keyphrases
- learning algorithm
- online learning
- reinforcement learning
- feature vectors
- learning process
- prior knowledge
- learning community
- learning problems
- mobile learning
- learning systems
- sufficient conditions
- learning phase
- preprocessing step
- multi layer
- knowledge acquisition
- learning scheme
- elementary school
- recurrent networks
- computer networks
- data sets
- supervised learning
- training data
- social networks
- machine learning
- neural network