AI & Machine Learning: From Scratch to Advanced Models

Early Access!

During Early Access, you can enroll at a reduced price and gain early access to course content—such as videos, documents, quizzes, and more—as it’s created. You’ll also have immediate access to the full course as soon as it’s complete.

Welcome!

We are excited to introduce you to the AI & Machine Learning: From Scratch to Advanced Models course, where you will go from understanding the fundamentals of AI to building and applying cutting-edge neural network architectures.

Whether you are an engineer, scientist, or AI enthusiast, this course will equip you with the skills to explore the immense ocean of possibilities that AI unlocks.

The course

This course is hands-on and practical, ensuring that you not only grasp the concepts but also apply them in real-world scenarios.

While many of our examples are from naval and ocean engineering, the principles and techniques are universal and relevant to any engineering or technical field. Plus, in the final project, you’ll have the opportunity to work on a dataset of your choice, making the learning experience truly personalized.

We’ll start by defining AI (Artificial Intelligence), Machine Learning, and Deep Learning, breaking down why these technologies are transforming industries and how they are shaping the future. We’ll explore why machine learning works mathematically and the different types of learning: supervised, unsupervised, and reinforcement learning.

Once we have a strong foundation, we’ll move into key topics such as datasets, features and labels, model training, validation and testing, the machine learning pipeline, overfitting and underfitting, evaluation metrics, hyperparameter tuning and optimization, and model deployment. We’ll work through core machine learning algorithms like Linear and Logistic Regression, Decision trees and Random Forest, K-Nearest Neighbors, and Support Vector Machines.

Once the fundamentals are clear, we’ll dive into Deep Learning, breaking down how neural networks function, training using gradient descent, backpropagation and autodifferentiation, and how to interpret what a neural network learns.
To solidify your understanding, you’ll implement a neural network from scratch, predicting ship resistance based on hydrodynamic data using only NumPy and Pandas.

Handling data correctly is essential for training powerful models. We’ll teach you data types, collection methods, and pre-processing: cleaning, normalization and feature engineering, and exploratory data analysis techniques. Then, we’ll transition to PyTorch, a widely used machine learning framework, and go through the entire machine learning pipeline from data preparation and model training to evaluation and optimization.

A great model is not just about building, it’s about fine-tuning for performance. We’ll cover evaluation metrics for classification and regression problems, handling overfitting and underfitting, and hyperparameter tuning techniques.

We won’t stop at basic neural networks: you’ll also explore convolutional neural networks for image processing, recurrent neural networks for sequential data analysis, and the latest trends and ethical considerations in AI and machine learning.

By the end, you will take on a capstone project where you can choose from a variety of datasets or work with the ShipD dataset from MIT to analyze 30,000 ship hull geometries and their wave resistance.

What will you learn?

By the end of this course, you will:

• have a deep understanding of how machine learning models work;
• be able to build, train, and evaluate your own AI models;
• have a working machine learning pipeline using real-world data;
• be prepared to explore more advanced AI topics.

More concretely, you will learn:

• AI, Machine Learning, and Deep Learning, why it’s booming, and how it applies to engineering.
• Key machine learning concepts: data handling, models, evaluation, and optimization.
• Core algorithms: linear regression, logistic regression, decision trees, K-nearest neighbors, and support vector machines, each with real-world applications.
• Deep Learning fundamentals: how neural networks learn; you’ll even build one from scratch.
• Implementation with PyTorch: data preprocessing, model training, evaluation, and optimization.
• Advanced architecture: convolutional neural networks (CNNs) and recurrent neural networks (RNNs).
• Practical insights and coding tips based on our hands-on experience working with large datasets and complex models.

Course Organization

The course is video-based and on-demand, allowing you to learn at your own pace, wherever and whenever you want.

It includes videos, quizzes, and downloadable documents and provides access to the course’s virtual private classroom, where you can interact with the instructor and other students.

Upon completing all lessons, passing the quizzes, and having your course assignment approved, you will receive the Course Certificate.

– Resources:

• Video lessons.
• English subtitles.
• Course Book (174 pages!)
• Quizzes.
• Final Assignment.
• Virtual Private Classroom.
• Course Certificate.

– Classroom:

– Prerequisites:

No prior knowledge of Artificial Intelligence (AI) or Machine Learning is required for this course. However, to fully benefit from the material, a basic understanding of Python programming, fundamental mathematical concepts, and a general familiarity with data analysis will be helpful.

To ensure you’re well-prepared, we’ve created the Introduction to Python course, which covers everything you need to know about Python—from the basics to more advanced topics like classes, with a focus on data analysis.

Although many of our examples are drawn from naval and ocean engineering, the principles and techniques taught in this course are applicable to any engineering or technical field.

A minimum Navalapp membership level of “Subscriber” (free membership) is required to enroll in this course.