AI-900 Question 101
Select 3A company wants to build an AI model to classify images of animals. They decide to use deep learning techniques. Which of the following are features of deep learning techniques that make them suitable for this task?
- A
They automatically extract features from raw data without requiring manual feature engineering.
- B
They require significantly less data compared to traditional machine learning algorithms.
- C
They leverage neural networks with multiple layers to learn complex patterns in data.
- D
They are specifically designed to work with structured tabular data.
- E
They are capable of handling unstructured data such as images, audio, and text.
Show answer and explanation
Correct answers: A, C, E
Explanation
Deep learning techniques are particularly suitable for tasks like image classification because they are capable of handling unstructured data, automatically extracting features, and leveraging deep neural networks to learn complex patterns. However, they typically require large amounts of data and are not specifically designed for structured data tasks.
- A. Correct.
Deep learning techniques can automatically extract features from raw data, which is a major advantage over traditional machine learning techniques that often require manual feature engineering.
- B. Incorrect.
Deep learning typically requires large amounts of data to perform well, especially for tasks such as image classification, making this statement incorrect.
- C. Correct.
Deep learning leverages neural networks with multiple layers (deep neural networks) to learn complex patterns and relationships in the data.
- D. Incorrect.
Deep learning is not specifically designed for structured tabular data; traditional machine learning algorithms often perform better for such data types.
- E. Correct.
Deep learning can handle unstructured data such as images, audio, and text effectively, which is why it is commonly used in tasks like image classification or natural language processing.