Elements of Machine Learning

Elements of Machine Learning

Announcements:

Homework

Homework 9 is available on the class homework page.

Communications

We will be using EdDiscussion to communicate, especially for Q&A about homework. The site is available through Sakai.

Lecture Recordings

Lectures are automatically recorded on the Panopto system, and are available here. You need a Duke affiliation to access the recordings.

Recitations

There will be no recitation sessions the first week of class.

Exams

The midterm exam is in person on Thursday, October 21 during the regular lecture period.

The final exam is in person and set by the Duke registrar for Saturday, December 11, starting at 7pm. It will be two hours in length.

Children learn in a variety of ways. Machine learning systems, on the other hand, mainly observe a multitude of examples and infer some pattern from them. This process is akin to data fitting, except that the goal is for the system to do well on new inputs, not just on the given examples. [Photo by Yannis H on Unsplash]

This course covers fundamental concepts of supervised machine learning, with sample algorithms and applications. "Supervised" means that for every example question given to the learner the corresponding correct answer is given as well. The course focuses on how to think about machine learning problems and solutions, rather than on a systematic coverage of techniques.

The class meets on Tuesdays and Thursdays from 8:30am to 9:45am in room 2231 of the French Family Science Building.

COMPSCI 371D, Duke University, Site based on the fluid 960 grid system