Post-Hoc Interpretation of Transformer Hyperparameters with Explainable Boosting Machines.
Kiron DebXuan ZhangKevin DuhPublished in: BlackboxNLP@EMNLP (2022)
Keyphrases
- hyperparameters
- post hoc
- model selection
- cross validation
- closed form
- random sampling
- bayesian inference
- support vector
- bayesian framework
- em algorithm
- gaussian process
- prior information
- noise level
- sample size
- maximum a posteriori
- gaussian processes
- maximum likelihood
- incomplete data
- feature selection
- incremental learning
- parameter settings
- learning algorithm
- fault diagnosis
- parameter space
- missing values
- expectation maximization
- grid search
- data mining
- high dimensional
- graph cuts
- markov random field
- active learning
- gaussian process regression
- prior knowledge