Deep Learning: Fundamentals and Applications
Introduction to the theory, architecture and application of artificial neural systems. Supervised, unsupervised and reinforcement learning in a single- and multiple-layer neural networks. Self-organizing maps. Learning capacity and generalization. Gradient-based learning methods and deep learning networks. Application to image recognition, pattern classification, nonlinear modeling and forecasting.
Prerequisites by Topic:
1. Linear algebra and matrix
2. Linear differential and difference equations
3. Mathematical Methods with Matlab
Textbook 1 (required): Fundamentals of Artificial Neural Networks, Mohamad H. Hassoun (The MIT Press, 1995).
Textbook 2 (required): Deep Learning, Goodfellow, Bengio and Courville (The MIT Press, 2016). Also, available free of charge in electronic form online.
MATLAB Student Version. You must have this software installed and working on your own laptop before the end of the first week of classes.
Faculty: Mohamad H. Hassoun,
Office: 3127 Engineering Bldg.
Office Hours: Wed. 2:30 – 4:30 and by appointment.
Final Project: 20%
Final Exam: 30%
Important note: All assignments and exams for this course are expected to be completed based on 100% individual effort. Any sharing of materials, program codes, etc. (relating to exams and assignments) among students is counted as cheating (refer to the Cheating Policy at the end of this document).
Percentage/Grade/(Honor Point Value)
95-100 A (4.00)
90-94 A- (3.67)
85-89 B+ (3.33)
80-84 B (3.00)
75-79 B- (2.67)
70-74 C+ (2.33)
65-69 C (2.00)
60-64 C- (1.67)
55-59 D+ (1.33)
50-54 D (1.00)
45-49 D- (0.67)
0-44 F (0.00)
Drop & Withdraw Deadlines:
Beginning the fifth week of class students are no longer allowed to drop but must withdraw from classes. It is the student's responsibility to request the withdrawal. Failure to do so will result in a grade of F. The withdrawal period for full-term classes ends at the end of the tenth week of the term. See the Academic Calendar for specific information on when the withdrawal period ends.
Attendance: Attendance is required for all lectures. A student will lose 1 percent (out of the total final score) for every unexcused absence from lectures and labs.
A grade of I will be available only if the student needs to complete at most the final exam.
The final exam is scheduled according to the published university final exam schedule.
Makeup Exams: Makeup exams are available only for students with documented emergencies.
Very Important: Your Professor is known to be very strict when it comes to attendance and deadlines. He expects you to conduct yourself as a professional. Here are few examples:
- He does not accept assignments submitted on the due date after he starts the lecture.
- If he sets a submission date for an assignment (say bonus problem) to be received by email before 5:00 pm on a certain day and you submit at 5:01pm then he would not accept your submission.
- Arrive more than 5 minutes late to class and you will be counted absent and lose points (1% of your final average is subtracted for each unexcused absence.
- In case of an emergency, you must email him about your absence before class and you must bring with you a legitimate documentation for your absence (e.g, doctor’s note, court note, etc.)
Turn in a hard copy of your assignment solution on the due date to your instructor five minutes before the beginning of the lecture. Arrive early because once the lecture starts your professor will stop accepting assignment solutions.
Note 1: All assignments must be completed based on individual effort. Any sharing (giving and/or receiving) of solutions, no matter how small, is considered cheating, and will lead to a zero grade for the assignment the second offence will lead to failing the course.
Note 2: All Matlab generated solutions, scripts, functions and plots must be included for each assignment. The solutions are preferred to be typed.
Note 3: All numerical calculations must be performed using Matlab or another language approved by your instructor. Copy and paste the calculations in your solution. Screen capture of Matlab graphs and results for inclusion in your assignments can be generated using the following free software: Free screen capture program (Jing).
Students with Disability: If you have a documented disability that requires accommodations, you will need to register with Student Disability Services for coordination of your academic accommodations. The Student Disability Services (SDS) office is located at 1600 David Adamany Undergraduate Library in the Student Academic Success Services department. SDS telephone number is 313-577-1851 or 313-577-3365 (TDD only). Once you have your accommodations in place, I will be glad to meet with you privately during my office hours to discuss your special needs. Student Disability Services’ mission is to assist the university in creating an accessible community where students with disabilities have an equal opportunity to fully participate in their educational experience at Wayne State University. You can learn more about the disability office at: www.studentdisability.wayne.edu
Cheating and Penalty for Cheating: Cheating is defined by the University as “intentionally using or attempting to use, or intentionally providing or attempting to provide, unauthorized materials, information, or assistance in any academic exercise.” This includes any group efforts on assignments or exams unless specifically approved by the professor for that assignment or exam. Evidence of fabrication or plagiarism, as defined by the University in its brochure “Academic Integrity,” will also result in downgrading for the course. Students who cheat on any assignment or during any examination will be assigned a failing grade for the course.
Prof. Hassoun’s policy on cheating:
- All work submitted for grading must be 100% individual effort (unless otherwise told beforehand by your professor as in the case of team assigned projects).
- The solutions to assignments (including any bonus problems and projects) might already be out there. Advice: Do not look at them, period!
- All work you submit for grading (assignments, exams, projects and bonus problems) must be 100% your own effort. You understand that once you submit your work for grading then you are automatically certifying that the work is 100% yours. Upon grading your work, if cheating is detected (no matter how small) on an Exam then you will FAIL the course. On all other graded work, the first cheating incidence (no matter how small) by a student will earn that student a zero for that piece of work. The second offence is an automatic failure of the course.
- And yes, your professor monitors website such as Chegg.com Freelancer.com and others.
The “Announcements” link is available for you to check important dates and announcements regarding this course. Please make sure you check it on a regular basis.