Machine Learning Fundamentals

Machine Learning Fundamentals

Utilizing machine learning to apply algorithms to their data has helped companies maximize efficiencies, pursue new markets, and create new products. This trend has prompted many industries to recognize the value of machine learning, creating a high demand for knowledge in this field. Understanding the theory of how machine learning algorithms work is not only important skill for being able to apply and debug code, but also an important skill for interviewing.

In this course, students will learn how machine learning algorithms work so they can better understand the strengths and weaknesses of popular machine learning algorithms and when to apply which algorithm in real world situations. Some of the algorithms we will cover in the course include logistic regression, k-nearest neighbors, decision trees, random forests, bagged trees, gradient boosting, principal component analysis, k-means, hierarchical clustering, support vector machines, naïve Bayes, and recommender systems. The course will also cover topics such as model validation, regularization, optimization functions, hyperparameter tuning, and methods to deal with unbalanced classes.

Toward the end of the course, we will touch on use cases where traditional machine learning algorithms are suboptimal and where deep learning can be more appropriate. These deep learning techniques will include image classification and natural language processing.

Course Topics:

• Classification and Regression
• How machine learning algorithms work
• How to tune algorithms
• Strengths and weaknesses of a variety of algorithms
• How to validate the performance of an algorithm

Learning Outcomes:

By the end of the course students will be able to:

• Understand the theory of how machine learning algorithms work
• Distinguish which algorithms are better fits for real world situations
• Be better prepared for the machine learning section of data science interviews
• Understand why deep learning is useful based on the limitations of traditional machine learning algorithms

Course typically offered: Quarterly, online

Prerequisites: Knowledge of linear algebra would be helpful, but not required.

Next steps: Upon completion, consider additional coursework in our specialized certificate in Machine Learning Methods to continue learning.

Contact: For more information about this course, please contact unex-techdata@ucsd.edu.

Course Number: CSE-41328
Credit: 3.00 unit(s)