Demystify Hyperparameters for Stochastic Optimization with Transferable Representations.
Jianhui SunMengdi HuaiKishlay JhaAidong ZhangPublished in: KDD (2022)
Keyphrases
- stochastic optimization
- hyperparameters
- model selection
- cross validation
- bayesian inference
- random sampling
- closed form
- support vector
- prior information
- noise level
- multistage
- bayesian framework
- gaussian process
- gaussian processes
- em algorithm
- sample size
- maximum a posteriori
- missing values
- parameter settings
- maximum likelihood
- incremental learning
- feature selection
- parameter space
- image reconstruction
- data sets
- machine learning