引導式測驗 · 中等 等級
機器學習測驗所需的數學
透過少量計算和對假設的檢查,將代數、矩陣維數、梯度和統計推理與具體的 ML 任務連結起來。
問題 1 的 8
For y = wx + b, with x = 3, w = 2 and b = 1, which prediction follows?
預覽每個問題8 問題
當您參加測驗時,就會出現答案和解釋。
- For y = wx + b, with x = 3, w = 2 and b = 1, which prediction follows?
- 7
- 6
- 9
- 3
- A learner changes w while keeping x and b fixed. What changes in the linear-prediction example?
- Only the identity of the input example
- The prediction rule’s coefficient
- The number of examples in the dataset
- The meaning of addition
- Two arrays have compatible multiplication dimensions. What still needs checking in an ML calculation?
- Nothing; compatible dimensions prove the model is correct
- Only the names of the files
- Whether features, their order and their units are appropriate
- Whether every value is positive
- Starting at 2 with gradient 0.6 and learning rate 0.1, what is the stated gradient-descent update?
- 2.6
- 2.06
- 1.4
- 1.94
- A training program successfully performs one gradient step. What does that demonstrate?
- That the step was performed, not that training converged or predictions are useful
- That the final model generalizes to every population
- That the loss function matches the user’s purpose
- That no more evaluation is needed
- Which mathematical idea helps a practitioner reason about sampling variation in evaluation?
- Only matrix transposition
- Probability and statistics
- Only the number of model layers
- The choice of a filename
- A model finds that two recorded variables are correlated. What additional claim does that alone fail to establish?
- That an association was observed
- That the dataset has at least two variables
- That changing one variable causes the other to change
- That the variables can be compared
- How should a learner choose the depth of mathematical study for an ML project?
- Require every advanced topic before trying any example
- Use the same course list for every role
- Study only what an interface displays
- Match depth to the task and check understanding with small examples