Machine Learning – Navalapp

Coming Soon!

Machine Learning (ML) is a specialized field within Artificial Intelligence (AI) that focuses on creating algorithms and statistical models enabling computers to learn and perform tasks without being explicitly programmed with fixed rules. Instead of adhering to a rigid set of instructions, ML systems identify patterns and learn from data, empowering them to make decisions, predictions, or classifications based on new, unseen data.

This course aims to provide a comprehensive introduction to machine learning with a particular focus on applications in naval engineering. By the end of this course, you will have a solid understanding of the foundational concepts and practical skills necessary to implement basic machine learning models in real-world naval engineering scenarios.

Course Content

The Course

Meet the Instructors

Introduction to the course

Introduction

What is Machine Learning?

Types of Machine Learning

Historical Context and Evolution of Machine Learning

Applications of Machine Learning in Naval Engineering

Importance of Machine Learning in Engineering

Fundamental Concepts

Features, Labels, and Datasets

Training, Validation and Testing Data

The Machine Learning Pipeline

Overfitting and Underfitting

Evaluation Metrics

Bias-Variance Trade-Off

Hyperparameters and Model Optimization

Model Deployment and Monitoring

Core Algorithms

Introduction to core algorithms

Linear Regression

Logistic Regression

K-Nearest Neighbors (KNN)

Support Vector Machines (SVM)

Deep Learning Basics

Introduction to Deep Learning

Neural Networks: Structure

Neural Networks: Training

Gradient Descent

Backpropagation

Autodifferentiation: Simplified MLP

Autodifferentiation: General MLP

Autodifferentiation: Dimensionality

Implementing a Simple Neural Network from Scratch

Data Handling and Preparation

Understanding Data: Types of Data and Data Collection Methods

Data Preprocessing: Cleaning and Normalization Techniques, Feature Engineering and Handling Missing Data

Data Preprocessing: Practical Session

Exploratory Data Analysis: Visualizing Data, Identifying Patterns and Anomalies

Exploratory Data Analysis: Practical Session

MLP Implementation with PyTorch

Introduction to the Problem

Setting Up the Environment and Ensuring Reproducibility

Loading and Preprocessing the Data

Building the Neural Network Model

Defining the Loss Function and Optimizer

Training the Model

Evaluating the Model

Saving and Loading the Model

Saving and Loading the Scalers

Conclusion and Further Exercises

Model Evaluation and Optimization

Evaluation Metrics for Classification Models (Accuracy, Precision, Recall and F1 Score) and Confusion Matrix

Evaluation Metrics for Regression Models (MAE, MSE, RMSE, R-score and MAPE)

Evaluation Metrics for Deep Learning Models (Log loss, Cross Entropy Loss and Perplexity)

Evaluation Metrics for Deep Learning Models: Practical Session

Overfitting vs Underfitting: Understanding Concepts and Techniques to Mitigate Overfitting

Overfitting vs Underfitting: Practical Session

Hyperparameter Tuning: Importance and Methods

Hyperparameter Tuning: Practical Session

Advanced Topics and Future Directions

Introduction to Advanced Architectures: Convolutional Neural Networks (CNNs)

Introduction to Advanced Architectures: Recurrent Neural Networks (RNNs)

Future Trends in ML and Naval Engineering: Emerging Technologies and Potential Future Applications

Final Assignment - Capstone Project

Project Proposal and Dataset Selection

Analyze and Clean the Dataset

Split the Dataset into Train, Validation, and Test

Normalize the Data

Define the Neural Network Architecture

Train the Neural Network

Test the Performance of the Neural Network

Predict the Residuary Resistance for a given Hull

Optional Extensions

Course Materials

Course Materials

Course Survey

Course Evaluation Survey

Summary

The Instructors

Andrés Pastor Sanchez

Javier Capel Jorquera

Antonio Medina Manuel