Understanding weight-magnitude hyperparameters in training binary networks.
Joris QuistYunqiang LiJan van GemertPublished in: CoRR (2023)
Keyphrases
- hyperparameters
- model selection
- cross validation
- bayesian inference
- support vector
- random sampling
- closed form
- grid search
- bayesian framework
- gaussian processes
- gaussian process
- prior information
- em algorithm
- maximum a posteriori
- noise level
- sample size
- maximum likelihood
- training set
- incremental learning
- incomplete data
- regularization parameter
- learning algorithm
- data mining
- graph cuts
- higher order