Artificial Intelligence (AI)：練習題

In Artificial Intelligence, which category of machine learning is defined by providing a computer with a dataset where the desired output (labels) is already known for each input?

Which of the following best describes the process of back propagation within an artificial neural network?

In the context of training an Artificial Neural Network (ANN), the back propagation of errors is a fundamental process. Which of the following statements best describes the mathematical mechanism used to update a weight \( w \) to minimize the error function \( E \)?

In the context of machine learning, which category describes a system that learns to perform a task through trial and error, receiving rewards or penalties based on the actions it takes within an environment?

Which machine learning category is characterized by an agent that learns to make decisions by performing actions in an environment to maximize a cumulative reward signal?

State the primary difference between supervised learning and unsupervised learning in terms of the training data provided to the system.

Describe the function of back propagation within an artificial neural network during the machine learning phase.

Artificial Intelligence (AI) systems often utilize machine learning to perform complex tasks such as speech recognition or data analysis.

(a) Explain what is meant by the term machine learning.

(b) Identify two characteristics that distinguish supervised learning from unsupervised learning.

(c) State one example of a real-world application that would typically use supervised learning.

Artificial Neural Networks (ANNs) are used in AI to model complex patterns in data and are fundamental to the field of deep learning.

(a) Describe the internal structure of a typical ANN, referring specifically to the different layers and how they are interconnected.

(b) In the context of training an ANN, explain the role of weights and how the back propagation of errors algorithm is used to improve the model accuracy.

(c) Explain why Deep Learning is considered a specialized subset of Machine Learning.