Arieyanna Jefferson's AI & ML Learning Booklet

The curriculum included key 6th grade math concepts from the Georgia Standards of Excellence. Topics included ratios and rates, fractions, decimals, integers, expressions, equations, geometry, and statistics.

POE, an AI-powered platform, was used to design interactive webpages for both learning and assessment. With POE, study guides were generated for each math topic, complete with explanations and visual examples.

The AI helped create digital practice quizzes and tests. These webpages allowed students to answer questions and receive instant feedback, making math practice more engaging and personalized.

By integrating AI with math content, students learned how technology can enhance understanding and retention. Creating and using these study and test tools also improved digital literacy and demonstrated the real-world application of AI in education.

The curriculum included key 6th grade math concepts from the Georgia Standards of Excellence. Topics included ratios and rates, fractions, decimals, integers, expressions, equations, geometry, and statistics.

POE, an AI-powered platform, was used to design interactive webpages for both learning and assessment. With POE, study guides were generated for each math topic, complete with explanations and visual examples.

The AI helped create digital practice quizzes and tests. These webpages allowed students to answer questions and receive instant feedback, making math practice more engaging and personalized.

By integrating AI with math content, students learned how technology can enhance understanding and retention. Creating and using these study and test tools also improved digital literacy and demonstrated the real-world application of AI in education.

Artificial Intelligence refers to computer systems designed to perform tasks that typically require human intelligence, such as understanding language, recognizing images, or making decisions.

AI has evolved from early rule-based systems in the 1950s to today's advanced learning models. Key milestones include the development of expert systems, neural networks, and modern deep learning breakthroughs.

Everyday AI includes tools like voice assistants (Siri, Alexa), movie recommendation systems, and self-driving cars, showing how AI is integrated into our lives.

As AI impacts society, important questions arise about privacy, fairness, job automation, and responsible use, making ethics a vital topic for all AI learners.

Early AI systems followed explicit rules and logic trees to solve problems, laying the foundation for more advanced learning systems.

Machine Learning (ML) allows computers to learn from data and improve their performance on specific tasks without being explicitly programmed for each decision.

AI is the broader field of creating intelligent systems, while ML is a subset where computers learn from examples and data.

ML models use data (features and labels) to discover patterns, build prediction rules, and make informed decisions.

Features are input variables; labels are the expected outputs. A dataset contains many examples, a model is trained on this data, and predictions are the model’s outputs.

Linear regression is an introductory ML model that finds a straight-line relationship between two numeric variables.

In supervised learning, models are trained using labeled data (where the correct answer is provided), such as image classification or spam detection.

Here, models work with unlabeled data to discover hidden patterns, such as grouping customers with similar preferences (clustering).

Linear regression predicts numbers, logistic regression predicts categories, and k-means clustering organizes data into similar groups without labels.

Practice building supervised models (like linear and logistic regression) and unsupervised models (like clustering) to understand both approaches.

JavaScript allows us to run ML models directly in the browser, making AI interactive and accessible without needing complex installations.

TensorFlow.js is a library for training and running ML models in the browser using JavaScript. It supports flexible, real-time applications.

You can quickly add TensorFlow.js to any web project using a Content Delivery Network (CDN), npm package, or direct script tags in HTML.

Basic programming concepts like variables, arrays, objects, functions, loops, and conditional statements are essential for using TensorFlow.js effectively.

Learn how to fetch datasets (like CSV files), process them, and visualize the results for interactive ML experiments.

Try out small code examples to load and view data, setting the stage for your own ML projects.

Linear regression helps predict a numeric output (like a car’s mpg) from one or more input features (like horsepower) by fitting a straight line to the data.

Logistic regression is used for classification, predicting whether something belongs to a category (such as spam or not spam).

Both linear and logistic regression are classic examples of supervised learning, requiring labeled datasets for training.

With TensorFlow.js, you can train models right in your browser, seeing results instantly and adjusting parameters on the fly.

Use TensorFlow.js to build and train regression models, applying them to real-world datasets like car horsepower and mpg.

Follow the TFJS Training Regression tutorial to learn step-by-step how to train and evaluate a model using the browser.

Neural networks are inspired by the human brain, using layers of nodes (neurons) to process data and recognize complex patterns.

Neural networks have layers (input, hidden, output), each with nodes that connect and pass signals, adjusted by weights during training.

Deep learning uses neural networks with many hidden layers, allowing them to solve more complex problems like image or speech recognition.

Convolutional Neural Networks (CNNs) are specialized for analyzing images, breaking them down into features like edges, shapes, and patterns.

CNNs are widely used in handwriting digit recognition, object detection, facial recognition, and many image-related AI tasks.

CNNs use convolutional layers to scan images, pooling layers to reduce data size, and activation functions to introduce decision-making.

Start with simple neural networks, then build CNNs for tasks like recognizing handwritten digits in images.

Apply your skills by building a model to predict car mpg from horsepower, experimenting with training and evaluation in the browser.

Follow the official TFJS regression tutorial to step through the process of loading data, defining a model, training, and visualizing results.

Use the MNIST dataset and TensorFlow.js to create a CNN that can identify handwritten digits with high accuracy.

Explore the official TFJS classification tutorial to learn how to set up your own digit recognition model.

Experiment with different model settings, visualize predictions, and understand model performance—all in your web browser!