Among the following training and evaluation strategies, which of them divides the dataset into k subsets and, at each iteration, a different subset is reserved for testing and all the others, used for training?

Question

BlackTom AI · Accepted Answer

Cross-validation divides the dataset into k equal (or nearly equal) subsets (folds) and, in k iterations, uses a different fold as the test set while using the remaining k−1 folds for training. This is k-fold cross-validation, a standard method to estimate model generalization.

1) Relevant concepts:
- k-fold cross-validation: partition data into k folds; perform k training/testing rounds.
- Leave-one-out: special case of cross-validation with k = n (each test fold has size 1).
- Holdout: a single fixed split (e.g., 70/30) with one training and one test set.
- Bootstrap: sampling with replacement to create training sets and using out-of-bag samples for testing.

2) Step-by-step reasoning:
- Choose k (common choices: 5, 10).
- Shuffle the dataset and split into k folds.
- For i from 1 to k: set fold i as test set, combine remaining folds as training set, train model, evaluate on fold i.
- Collect the k performance measures and average them to get the cross-validated estimate.

3) Example calculation:
- Dataset size N = 100, choose k = 5 → fold size = 20.
- Iteration 1: train on 80, test on 20 → accuracy1 = 0.82.
- Iteration 2: train on other 80, test on 20 → accuracy2 = 0.79.
- ... Iteration 5: accuracy5 = 0.81.
- Average accuracy = (0.82 + 0.79 + ... + 0.81) / 5 (compute sum then divide by 5) → final estimate.

4) Why this is correct:
The question explicitly describes reserving a different subset for testing at each iteration while training on the rest — exactly the k-fold cross-validation procedure.

5) Why other options are incorrect:
- Bootstrap: uses sampling with replacement to form many training sets; test estimates use out-of-bag samples, not systematic k-fold partitions.
- Holdout: uses a single fixed split (one test set), not k iterative folds.
- Leave one out: while conceptually similar, it is the special case k = N (each fold = 1 sample); the general description of "k subsets" refers to k-fold cross-validation.

6) Conclusion:
The correct answer is Cross Validation, matching the provided answer.

Among the following training and evaluation strategies, which of them divides the dataset into k subsets and, at each iteration, a different subset is reserved for testing and all the others, used for training?单项选择题

类似问题

What is the purpose of cross validation?

What is the purpose of performing cross-validation?

When training models using machine learning, a very powerful technique for tuning and optimizing the best model is:

Why was model-building strategy #1 used to estimate generalization performance on the income data for the decision tree, logistic regression, and random forest models?

In k-fold cross-validation, what happens in each iteration?

What is one advantage of k-fold cross-validation over a single holdout set?

Which of the following is TRUE about cross validation?

更多留学生实用工具

智能学习助手

风格化写作助手

论文查重助手

文献引用助手

课堂转译助手

课堂笔记助手

Quiz搜索助手

学校历年真题

智能刷题助手

智能匹配练习

希望你的学习变得更简单