Tensor Programs V: Tuning Large Neural Networks via Zero-Shot Hyperparameter Transfer.
Greg YangEdward J. HuIgor BabuschkinSzymon SidorXiaodong LiuDavid FarhiNick RyderJakub PachockiWeizhu ChenJianfeng GaoPublished in: CoRR (2022)
Keyphrases
- neural network
- pattern recognition
- artificial neural networks
- high order
- gaussian processes
- higher order
- back propagation
- cross validation
- hyperparameters
- multilayer perceptron
- tensor product
- fuzzy logic
- fuzzy systems
- knowledge transfer
- multi layer
- neural nets
- diffusion tensor
- object detection
- computer programs
- tuning parameters
- maximum a posteriori
- training process
- transfer learning
- fault diagnosis
- prior knowledge
- machine learning