Supervised Learning

Narrow AI

General AI

Deep Learning

To eliminate the need for human creativity.

To create more complex problems for humans.

To replace all human jobs.

To perform tasks that require human intelligence.

Herbert Simon

Alan Turing

Marvin Minsky

John McCarthy

1950s

1980s

1970s

1960s

Analyzing historical data for trends.

Creating new content based on learned patterns from existing data.

Performing complex mathematical calculations.

Translating text from one language to another.

GPT-3

FastText

Word2Vec

BERT

Deep learning is broader; machine learning is a subset using decision trees.

Machine learning requires more data than deep learning.

Deep learning can only be used for image processing tasks.

Machine learning is broader; deep learning is a subset using neural networks.

Image compression

Network security enhancement

Predictive analytics

Data storage optimization

Support Vector Machine

K-means Clustering

Principal Component Analysis

Linear Regression

Training data is the final output of a machine learning model.

Training data is the data collected after model deployment.

Training data is used for testing the performance of a model.

Training data is the dataset used to train a machine learning model.

To eliminate the need for data preprocessing.

The purpose of neural networks in deep learning is to model complex patterns and relationships in data for tasks such as classification, regression, and feature extraction.

To simplify data storage and retrieval.

To replace traditional programming methods entirely.

Reliance on handcrafted features

Use of shallow neural networks

Limited to linear models

Use of deep neural networks with multiple layers

Reinforcement learning is a type of machine learning that focuses solely on data analysis.

Reinforcement learning is a type of machine learning focused on training agents to make decisions through trial and error to maximize rewards.

Reinforcement learning is a technique for clustering data into groups.

Reinforcement learning is a method for supervised learning using labeled data.

The Turing Test assesses a machine's ability to exhibit human-like intelligence.

The Turing Test determines a machine's ability to perform calculations.

The Turing Test evaluates a machine's physical appearance.

The Turing Test measures a machine's speed in processing data.

Statistical analysis

Supervised learning, Unsupervised learning, Reinforcement learning, and Deep learning are types of machine learning, while 'Data mining' is NOT a type of machine learning.

Predictive modeling

Data analysis

Data is only useful for storage purposes.

Data has no impact on AI model accuracy.

Data is irrelevant in the AI development process.

Data is crucial for training AI models and improving their performance.

Overly complex algorithms

Lack of quality data

Insufficient funding

Lack of interest from stakeholders

Generative AI creates content by randomly selecting words from a dictionary.

Generative AI generates new content by copying existing works without modification.

Generative AI creates new content by learning patterns from existing data and generating new samples based on those patterns.

Generative AI creates content by using pre-defined templates without learning from data.

AI will only be used for grading exams.

AI will have no impact on student engagement.

The future potential of AI in education includes personalized learning, real-time feedback, automation of administrative tasks, and enhanced accessibility.

AI will replace teachers entirely.