AI News, What is it like to take 6.867 (Machine Learning) at MIT?

What is it like to take 6.867 (Machine Learning) at MIT?

But, since most of the people taking the class are either graduate students who already know the material or undergraduates who have taken 6.036 (introduction to machine learning), the class does move pretty quickly.

Throughout the semester, you basically derive the most common machine learning models (ridge regression, logistic regression, SVMs with kernels, neural networks, graphical models, EM, Gaussian mixture models) and really dig into the mathematics of why they work.

The graded homework is three academic-style papers, in which you implement a specific algorithm derived in class, investigate how various parameters affect the results, and apply it to some interesting real-world data set.

For example, my project was trying to classify facial expressions using a variety of different techniques, including SVMs and neural networks based on facial landmarks and a deep convolutional neural net.

Lecture 10: Neural Machine Translation and Models with Attention

Lecture 10 introduces translation, machine translation, and neural machine translation. Google's new NMT is highlighted followed by sequence models with ...

Ellen's Helping Out with Homework!

Is there anything she can't do? Ellen offered to help her viewers with their homework. This is how it turned out!

School started ! Elsa and Anna toddlers - first day - new students - Barbie is teacher - classroom

In this toys dolls parody video, toddlers Anna and Elsa are back to classroom, the school has begun ! They meet new students and make new friends ! Watch the ...

How To Learn Faster

Get smart with Brilliant: Subscribe: The 9 BEST Scientific Study Tips: Created

13. Classification

MIT 6.0002 Introduction to Computational Thinking and Data Science, Fall 2016 View the complete course: Instructor: John Guttag ..

Muscles, part 1 - Muscle Cells: Crash Course A&P #21

We're kicking off our exploration of muscles with a look at the complex and important relationship between actin and myosin. Your smooth, cardiac, and skeletal ...

Andrew Trask - Really Quick Questions with an AI Researcher

I ask 67 questions to Oxford Scholar and AI researcher Andrew Trask as we go for a walk through Granary Square in London, England. Trask is a PhD student at ...

Barbara Oakley: "Learning How to Learn" | Talks at Google

About the Book: Whether you are a student struggling to fulfill a math or science requirement, or you are embarking on a career change that requires a higher ...

1. Introduction and Scope

MIT 6.034 Artificial Intelligence, Fall 2010 View the complete course: Instructor: Patrick Winston In this lecture, Prof. Winston ..

Lecture 3 | Loss Functions and Optimization

Lecture 3 continues our discussion of linear classifiers. We introduce the idea of a loss function to quantify our unhappiness with a model's predictions, and ...