Skip to content
Course materials for CSCI 378: Deep Learning (Reed College)
Branch: master
Clone or download
Fetching latest commit…
Cannot retrieve the latest commit at this time.
Permalink
Type Name Latest commit message Commit time
Failed to load latest commit information.
admin
code/rubik
hw
notes
.gitignore
README.md

README.md

CSCI 378: Deep Learning

Spring 2019 Syllabus

Basic Information

Professor: Mark Hopkins, hopkinsm@reed.edu

Class Schedule: MWF 310-4pm in Physics 240A

Office Hours: MW 4:10-6pm, Th 10am-noon in Library 314 (M 4:10-5 and Th 10-11am are reserved for this class; the remainder are shared with CSCI 121).

Textbook (optional): Deep Learning by Ian Goodfellow, Yoshua Bengio, and Aaron Courville.

Website: http://markandrewhopkins.com/csci-378-deep-learning

Overview

This course teaches you the fundamentals you need, in order to be an informed, well-rounded practitioner of deep learning. It will proceed through three phases:

  1. The Pool: In which we learn/review the fundamental background for deep learning, including gradient descent, simple regression models, important probability distributions, regularization, and matrix manipulation (with the Python torch package).
  2. The Shallows: In which we learn about the most mainstream concepts in deep learning, including multilayer feedforward networks, convolutional neural networks, and recurrent neural networks.
  3. The Abyss: In which we go into the cutting edge, including advanced neural architectures, open research questions, and important NLP and computer vision applications.

Coursework

Homework: There will be short but very regular (i.e. most class sessions) homework assignments. They are important to do, so that you can learn the material well. Even if you can't get to the solution, please try to hand in a good faith effort. We will spend the first section of each class tackling the homework assignment from the last class. You will be required to hand in homework solutions, with the freedom to skip four homeworks over the course of the semester without penalty. I would, however, encourage you not to exercise this freedom unless necessary.

Projects: There will be an ongoing sequence of projects during the course, curated specifically for this course offering.

Exams: There will be three exams during the course: two midterms and a final. Each exam is weighted equally and covers one phase of the course. However, the material in the course builds on the previous material, so while the final will be focused on the final third of the course, you will probably need a solid understanding of the first two thirds in order to do well.

Collaboration

Collaborating on homework and projects is permitted, but each student must write up homework independently, and must do the actual programming on the projects independently (no cutting and pasting somebody else’s code!) Also, you should acknowledge the names of anyone who you collaborated with.

Reading Assignments

Reading assignments will be posted on the website a minimum of two days in advance of each lecture. I will assume that the reading is done prior to lecture.

Disability Accommodation

If you have a disability for which you are or may be requesting an accommodation, you are encouraged to contact both your professor and the Office of Disability Support Services, disability-services@reed.edu or 503-517-7921 as early as possible in the semester. Please be aware that requests may take several weeks to implement once approved, and that accommodations are not retroactive.

You can’t perform that action at this time.