Which of the following statements about gradient descent and learning rate is true?

Question

BlackTom AI · Accepted Answer

To evaluate the statements, we need to recall how gradient descent and learning rate function in optimization.
Option 1: 'Gradient descent is only used in unsupervised learning algorithms.' This is incorrect because gradient descent is a general optimization method used across many learning paradigms, including supervised learning (e.g., training neural networks with labeled data), unsupervised learning, and reinforcement learning. Limiting it to unsupervised learning misrepresents its applicability.
Option 2: 'The learning rate controls the size of the steps taken during the optimization process.' This is accurate. The learning rate determines how far we move along the negative gradient (the direction of steepest descent) in each iteration. A larger learning rate means bigger steps, while a smaller rate yields finer, more gradual updates.
Option 3: 'A very large learning rate guarantees faster and more accurate convergence.' This is misleading. While a larger learning rate can reduce the number of iterations, it does not guarantee faster or more accurate convergence; in fact, it often causes unstable training or divergence if the steps overshoot the minimum.
Option 4: 'Gradient descent updates model parameters by moving in the direction of increasing loss.' This is incorrect. Gradient descent moves parameters in the direction that minimizes loss, i.e., in the opposite direction of the gradient of the loss function with respect to the parameters.

BU.330.775.T2.FA25 Final- Requires Respondus LockDown Browser

Which of the following statements about gradient descent and learning rate is true?

查看解析

登录即可查看完整答案

类似问题

Which statement is correct?

在梯度下降中如何更新参数？How do we update the parameters in gradient descent?

Which of the following best describes the role of the gradient in gradient descent?

Question at position 20 What is gradient descent?A way to calculate the slope of a loss function A regularization techniqueA method to normalize network activationsAn optimization algorithm to find local minimum of a function

Which statement best describes the use of gradient descent in linear classification?

How does random initialization help in gradient descent methods?

更多留学生实用工具

考试浏览器助手

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