Listen to this story
|
Last month, Andrew Ng’s DeepLearning.AI introduced a new course, “Mathematics for Machine Learning and Data Science Specialisation.” The same course is available on Coursera, led by Luis Serrano and co-created by Ng alongside Anshuman Singh, Magdalena Bouza, and Elena Sanina. This new course gives people an intuitive understanding of AI’s most important maths concepts.
Enrol for the course for free here.
“I have often said ‘don’t worry about it’ when it comes to maths because maths shouldn’t hold anyone back from making progress in ML,” said Ng. He said that understanding some key topics in linear algebra, calculus, probability, and stats will help you better get learning algorithms to work.
Further, he said this specialisation was designed with various interactive visualisations to help people see how maths works. “Maths is not about memorising formulas; it is about sharpening your intuition,” added Ng.
What to Expect?
This is a beginner-friendly programme where people master the fundamentals of mathematics of machine learning. This specialised course is said to use innovative mathematics pedagogy that helps students learn quickly and properly, with courses that use easy-to-follow plugins and visualisations to help students learn how maths behind ML works.
Once completed, students will understand the maths behind the algorithms and data analysis techniques. Plus the know-how to implement them into their machine learning career.
Outcome:
Andrew Ng and the team believe that by the end of this course, students will be able to represent data as vectors and matrices and spot their properties using singularity, rank and linear independence concepts. Also, they will be able to apply common vector and matrix algebra operations like the dot product, inverse and determinants.
In addition to this, students will be able to express certain types of matrix operations as linear transformations, alongside applying concepts of eigenvalues and eigenvectors to machine learning problems, optimise different types of functions, perform gradient descent in neural networks with various activation and cost functions, describe and quantify the uncertainty inherent in predictions/forecasting made by ML and data science, apply common statistical methods like MAP and MLE and more.