Implicit Acceleration and Feature Learning in Infinitely Wide Neural Networks with Bottlenecks.
Etai LittwinOmid SaremiShuangfei ZhaiVimal ThilakHanlin GohJoshua M. SusskindGreg YangPublished in: CoRR (2021)
Keyphrases
- neural network
- learning process
- learning algorithm
- learning systems
- supervised learning
- cascade correlation
- error back propagation
- pattern recognition
- neural nets
- learning problems
- online learning
- learning tasks
- fuzzy logic
- hidden layer
- back propagation
- knowledge acquisition
- mobile robot
- active learning
- prior knowledge
- artificial neural networks
- support vector
- genetic algorithm
- machine learning