Тест: Introduction to Machine Learning — 10 въпроса

1. What is machine learning primarily defined as?

A. A technique where computers are only able to perform tasks they are explicitly programmed for.

B. A process of manually coding rules for decision-making in computers.

C. A subset of artificial intelligence that enables computers to learn from data patterns and make decisions or predictions without explicit programming.

D. A method where computers are explicitly programmed for each task.

Answer: (C)

Обяснение

Machine learning is defined as a subset of artificial intelligence that enables computers to learn from data patterns and make decisions or predictions without explicit programming. This distinguishes it from traditional programming, where explicit instructions are provided for each task.

2. What is the primary purpose of an algorithm in machine learning?

A. To analyze data and identify patterns.

B. To store data for future use.

C. To manually classify data points.

D. To replace human decision-making completely.

Answer: (A)

Обяснение

An algorithm in machine learning provides a step-by-step procedure for analyzing data and discovering patterns, which is central to how models learn from data.

3. Who developed the Perceptron, an early neural network model, in 1957?

A. Alan Turing

B. Frank Rosenblatt

C. Geoffrey Hinton

D. Yann LeCun

Answer: (B)

Обяснение

The Perceptron was developed by Frank Rosenblatt in 1957. It was one of the earliest neural network models and marked a significant milestone in the history of machine learning. Alan Turing proposed the Turing Test in 1950, Geoffrey Hinton contributed to backpropagation in 1986, and Yann LeCun is known for his work on convolutional neural networks, but not for developing the Perceptron.

4. Who developed the perceptron, one of the earliest neural network models, in 1957?

A. Geoffrey Hinton.

B. Frank Rosenblatt.

C. Alan Turing.

D. Yann LeCun.

Answer: (B)

Обяснение

Frank Rosenblatt developed the perceptron in 1957, marking an important milestone in neural network history.

5. What is the primary role of supervised learning in machine learning?

A. To reduce the dimensionality of data for visualization

B. To optimize decision-making through trial-and-error interactions

C. To enable models to learn from labeled data for prediction or classification

D. To discover hidden patterns in unlabeled data

Answer: (C)

Обяснение

Supervised learning's main purpose is to train models using labeled data so they can predict outputs for new inputs, making it essential for tasks like classification and regression.

6. Which milestone in machine learning occurred in 2012 with the success of AlexNet?

A. The invention of the perceptron.

B. The advent of deep learning.

C. The introduction of the Turing Test.

D. The creation of support vector machines.

Answer: (B)

Обяснение

In 2012, AlexNet's success marked the breakthrough of deep learning, a subset of neural networks with many layers.

7. What does 'features' refer to in a machine learning context?

A. The output variable the model predicts.

B. Input variables or attributes used to make predictions.

C. The entirety of training data.

D. The algorithm used to analyze data.

Answer: (B)

Обяснение

Features are input variables or attributes, such as age or income, used by models to make predictions or classifications.

8. What is a key difference between supervised and unsupervised learning?

A. Supervised learning uses labeled data; unsupervised uses unlabeled data.

B. Supervised learning is only for classification; unsupervised is only for regression.

C. Supervised learning does not require data; unsupervised does.

D. Unsupervised learning always involves neural networks.

Answer: (A)

Обяснение

Supervised learning trains models on labeled data to predict specific outputs, while unsupervised learning finds patterns in unlabeled data.

9. Which of the following best describes the concept of overfitting in machine learning?

A. When a model performs well on training data but poorly on new data.

B. When a model cannot learn from training data.

C. When a model underestimates the complexity of the data.

D. When a model is too simple to capture data patterns.

Answer: (A)

Обяснение

Overfitting occurs when a model learns the training data too closely, including noise, leading to poor generalization on new, unseen data.

10. Why is the development of deep learning considered a milestone in machine learning history?

A. It introduced the use of simple linear models.

B. It enabled neural networks with many layers to perform complex tasks successfully.

C. It replaced all other machine learning techniques.

D. It was the first time machines could learn from data.

Answer: (B)

Обяснение

Deep learning involves neural networks with multiple layers that can model complex patterns, leading to breakthroughs in tasks like image recognition.