Too-Hot-to-Handle: Insights into Temperature and Noise Hyperparameters for Differentiable Neural-Architecture-Searches.
Joschua ConradSimon WilhelmstätterRohan AsthanaVasileios BelagiannisMaurits OrtmannsPublished in: AICAS (2024)
Keyphrases
- hyperparameters
- neural architecture
- noise level
- model selection
- neural network
- cross validation
- closed form
- bayesian framework
- feed forward
- bayesian inference
- noise reduction
- random sampling
- noisy images
- prior information
- maximum a posteriori
- em algorithm
- sample size
- maximum likelihood
- basis functions
- support vector
- multilayer perceptron
- missing data
- activation function
- missing values
- incremental learning
- artificial neural networks
- incomplete data
- error rate
- multi layer perceptron