/ml-master-course-ai-university

Course created by the model itself, with several iterations

Primary LanguageJupyter Notebook

Virtual Machine Learning Master Course

Initial Plan:

1: Foundations and Core Concepts

1.1: Foundations

  • Weeks 1-4: Introduction to Machine Learning, Python basics.
  • Weeks 5-8: Probability, Statistics, and Data Analysis.
  • Weeks 9-12: Algorithms and Data Structures for ML.
  • Weeks 13-16: Introductory Projects and Hands-on Python Exercises.

2.2: Core ML Concepts

  • Weeks 1-4: Supervised and Unsupervised Learning Techniques.
  • Weeks 5-8: Introduction to Neural Networks and Deep Learning.
  • Weeks 9-12: Feature Engineering and Data Preprocessing.
  • Weeks 13-16: Semester Project on a Supervised Learning Problem.

2: Advanced Topics and Specialization

  • Weeks 1-4: Natural Language Processing and Computer Vision.
  • Weeks 5-8: Reinforcement Learning and Predictive Analytics.
  • Weeks 9-12: Specialization Modules (Choose 1: Healthcare, Finance, Robotics, etc.).
  • Weeks 13-16: Advanced Projects in chosen specialization.

2.1: Real-World Applications and Capstone Project

  • Weeks 1-4: Ethics in AI, Bias and Fairness.
  • Weeks 5-8: Large Scale Machine Learning and Cloud Computing.
  • Weeks 9-16: Capstone Project: From Proposal to Deployment.

3: Assessments and Projects

  • Bi-weekly quizzes and coding assignments.
  • End of each semester: Comprehensive Take-Home Exams.
  • Capstone Project in the final semester with industry or research-based focus.

4: Workshops and Guest Lectures

  • Regular workshops on the latest ML tools and technologies.
  • Guest lectures from industry experts and renowned researchers.

5: For a top-tier program like ours, let's set it up like this:

  • 3 Lectures per Week: Each lasting about 2 hours, focusing on core concepts, theory, and applications.
  • 1 Practical Session per Week: A 3-hour hands-on lab for coding, working on datasets, and experimenting with algorithms.
  • 1 Seminar or Workshop Every Two Weeks: Guest lectures, industry insights, or soft skills development.

1: Getting Started with Machine Learning and Python

Introduction to Machine Learning

Topic: Overview of ML, history, key concepts.
Material:

  • Video: "What is Machine Learning?" for a basic overview.
  • Video: "The 7 Steps of Machine Learning" to understand the ML process.
  • TensorFlow Playground: https://playground.tensorflow.org/
  • Activity: Discussion on ML's impact in various industries.

Python Programming Basics

Topic: Python syntax, basic data types, control structures.
Material:

Python for Data Science

Topic: Introduction to NumPy and Pandas, basic data manipulation.
Material:

Assignments:

  • Reading: Chapters from "Python for Data Analysis" by Wes McKinney (e-book or online resource).
  • Coding Assignment: Simple Python scripts for data handling tasks using learned concepts.

Additional Resources:

2: Diving Deeper into Machine Learning Concepts

Class 4: Introduction to Probability and Statistics

Topic: Basic probability theory, descriptive statistics, and inferential statistics.
Material:

Class 5: Calculus and Linear Algebra in Machine Learning

Topic: Algebra Material:

Class 6: Neural Net Structure

Topic: Structure of Neural-Nets Material:

Topic: Activation Functions Material:

  • sigmoid, sigmoid gradient
  • hyperbolic-tangent
  • ReLU
  • softmax

Topic: Transformers Material:

  • video: What are Transformers?
  • Use case: Word Embedding
    • self attention Paper
      • encoder. decoder, position encoders

Class 7: Back Propagation & Gradient Descent

Topic: Backpropagation introduction Material:

Topic: Understanding Gradient Descent Material:

Topic: Measure Performance of Training Material:

3: HANDS ON: build a Neuron Net from scratch

Class 8: Practical - Build a Neural Net from Scratch

Material:

4: HANDS ON: build a Neuron Net from scratch + Learning Algorithms

Class 8 continuation: Practical - Build a Neural Net from Scratch

Material:

Class 9: Genetic Algorithms

Material:

  • research..... NEAT - Neuro Evolution of Augmented Topology
  • research..... PPO - Proximal Policy Optimization

5: HANDS ON: build DeepNeuralNet with Pytorch and test with a game

Class 10: Pytorch 25hrs Full Course

Material:

Class 11: Convolutional NN Architectures

Material: