Asymptotic Equivalence of Bayes Cross Validation and Widely Applicable Information Criterion in Singular Learning Theory

By Sumio Watanabe. 2010

2,808 citations recorded by the source corpus.

In regular statistical models, the leave-one-out cross-validation is asymptotically equivalent to the Akaike information criterion. However, since many learning machines are singular statistical models, the asymptotic behavior of the cross-validation remains unknown. In previous studies, we established the singular learning theory and proposed a widely applicable information criterion, the expectation value of which is asymptotically equal to the average Bayes generalization loss. In the present paper, we theoretically compare the Bayes cross-validation loss and the widely applicable information criterion and prove two theorems. First, the Bayes cross-validation loss is asymptotically equivalent to the widely applicable information criterion as a random variable. Therefore, model selection and hyperparameter optimization using these two values are asymptotically equivalent. Second, the sum of the Bayes generalization error and the Bayes cross-validation error is asymptotically equal to 2λ/n, where λ is the real log canonical threshold and n is the number of training samples. Therefore the relation between the cross-validation error and the generalization error is determined by the algebraic geometrical structure of a learning machine. We also clarify that the deviance information criteria are different from the Bayes cross-validation and the widely applicable information criterion.

Subfields: Artificial Intelligence, Probabilistic & Bayesian Methods

How researchers see this paper

The researchers whose trusted networks rank it highest, strongest advocate first. Every researcher gets a different order.

In Artificial Intelligence

  1. Aki Vehtari
  2. Andrew Gelman
  3. Quoc V. Le
  4. Kyunghyun Cho
  5. Mohit Bansal
  6. Yoshua Bengio
  7. Sergey Levine
  8. Percy Liang

In Probabilistic & Bayesian Methods

  1. Aki Vehtari
  2. Andrew Gelman
  3. Yong Luo
  4. Khaleel Al-Harbi
  5. Yuling Yao
  6. J. Guillochon
  7. Daniel P. Simpson
  8. José Miguel Hernández-Lobato

Citations in the corpus