Syllabus

Course Description

Deep Learning is an advanced application of neural networks within the field of machine learning. This course motivates and covers the foundations of deep learning, alongside its current technological applications. This course will provide the knowledge and skillset to navigate the current landscape of artificial intelligence algorithms employed across various fields of study.

Course Objectives

To give an overview of the foundations and recent advances in neural net algorithms. The first 2/3 of the course focuses on supervised learning (“learning with a teacher”). The last 1/3 focuses on unsupervised learning (“learning without a teacher”) and reinforcement learning.

This course will be taught as a “flipped classroom”, where you will study and review the posted notes, slides, papers, and videos prior to the once-a-week discussion section that we will have. During this discussion section, we will go over problems and concepts that you will need to understand exam material and submitted homework.

Course Prerequisites

The prerequisites are: MATH 152 - Calculus II, STAT 380 - Statistics for Applications

Assessment Criteria

The evaluations comprising a student’s grade in the course are:

4 HW’s: 40% (10% each)
Group Project: 40%
Midterm Exam: 20%

Grading Scale

The grading scale (curve included) for the course is:

A ≥ 80%; B ≥ 65%; C ≥ 55%; D ≥ 50%

Topics Covered

Linear Models
Multilayer Perceptrons
Backpropagation
Distributed Representations
Automatic Differentiation
Optimization
Convolutional Networks
Image Classification
Generalization
Recurrent Neural Nets
Exploding and Vanishing Gradients
Autoregressive and Reversible Models
Variational Autoencoders
Generative Adversarial Nets
Bayesian Neural Nets

Late or missed assignments/exams

No makeup exams or assignments will be given. If you miss an exam or assignment, there must be documentation in writing, provided to me or to the department administrative assistant. I will still review whether you have exercised due diligence. Otherwise, you will receive 0 points for the missed exam or assignment.

Textbooks

There are no required textbooks for this course. The necessary reading material will be provided on the course website.

Academic Honesty

Plagiarism is the use of another person’s words or ideas without crediting that person. Plagiarism and cheating will not be tolerated and may lead to failure on an assignment, in the class, or dismissal from the University, per the SIUE academic dishonesty policy. Students are responsible for complying with University policies about academic honesty as stated in the University’s Student Academic Conduct Code.

Academic Integrity

Students are reminded that the expectations and academic standards outlined in the Student Academic Code (3C2) apply to all courses, field experiences and educational experiences at the University, regardless of modality or location. The full text of the policy can be found here: Student Academic Code.

Recordings of Class Content

Faculty recordings of lectures and/or other course materials are meant to facilitate student learning and to help facilitate a student catching up who has missed class due to illness. As such, students are reminded that the recording, as well as replicating or sharing of any course content and/or course materials without the express permission of the instructor of record, is not permitted, and may be considered a violation of the University’s Student Conduct Code (3C1), linked here: Student Conduct Code.

Accessibility

Students needing accommodations because of medical diagnosis or major life impairment will need to register with Accessible Campus Community & Equitable Student Support (ACCESS) and complete an intake process before accommodations will be given. The ACCESS office is located in the Student Success Center, Room 1270. You can also reach the office by e-mail at myaccess@siue.edu or by calling (618) 650-3726. For more information on policies, procedures, or necessary forms, please visit the ACCESS website at www.siue.edu/access.

Potential for Changes in Course Schedule or Modality

As the COVID-19 pandemic continues, there remains a possibility that planned classroom activities will need to be adjusted. Depending on circumstances and following state-issued recommendations, potential changes include changes in course modality (e.g., transition from face-to-face to online) or in course scheduled meetings. These changes would be implemented to ensure the successful completion of the course. In these cases, students will be provided with an addendum to the class syllabus that will supersede the original version.

At the discretion of the instructor, all material, assignments, and deadlines are subject to change with prior notice. It is your responsibility to stay in touch with your instructor, review the course site regularly, or communicate with other students, to adjust as needed if assignments or due dates change.