Learning AI Through Visualization

By the end of this course, learners will be able to:

• Understand the roles and relationships between artificial intelligence, machine learning, and deep learning, and how each field contributes to modern technological advancements.
• Understand the importance of dataset specification and model choice, recognizing their impact on AI model performance and decision-making processes.
• Develop a solid understanding of various optimization routines, including brute-force search, gradient-free methods, and gradient descent, for improving model accuracy.
• Construct and optimize linear regression models using different approaches such as batch, mini-batch, and stochastic gradient descent, and analyze error functions.
• Familiarize yourself with major deep learning architectures and activation functions, including ReLU, sigmoid, and softmax, and their use in transforming data.
• Gain insights of generative adversarial networks (GANs) and diffusion models, understanding their use in image generation, text completion, and creative applications in AI.

Module 1: Generic Recipe for AI-Related Problems

This module covers the evolution of AI and the essential components of becoming a proficient AI chef.

Module 2: Visual Introduction to Optimization

This module visually explores optimization methods essential for effective machine learning and understanding model performance.

Module 3: Visual Introduction to Machine Learning through Linear Regression

This module introduces machine learning concepts, focusing on linear regression, optimization, and gradient descent techniques.

Module 4: Visual Introduction to Deep Learning

This module covers deep learning fundamentals, nonlinear feature transformations, and neural network architectures, with a focus on optimization.

Module 5 Part 1: Overview and Evolution of Language Models

This part covers neural architectures and approaches in sequence analysis and comprehension, focusing on neural networks and parsing techniques.

Module 5 Part 2: Sequence Analysis - Neural Network-Based Approaches

This part introduces neural architectures like convolutional networks, autoencoders, recurrent networks, and LSTM for advanced language models.

Module 5 Part 3: Large Language Models (LLMs) - Quick Overview

This part focuses on large language models, their architecture, limitations, and evaluation, including issues with GPT-based models.

Module 6 Part 1: Introduction to Generative AI (GenAI) Part I

This part introduces generative AI (Gen AI), its early development, applications, and the core technology behind it—generative adversarial networks (GANs).

Module 6 Part 2: Introduction to Generative AI (GenAI) Part II

This part explores the variations of GANs, their applications in finance, and introduces diffusion models and their impact.