Statistics 654: Applied Deep Learning (2 units)
Department of Statistics and Biostatistics, CSU East Bay
Syllabus
Lecture:
- Section 1: MW noon to 1:40, North Science, Rm. N112
Instructor:
Prof. Eric A. Suess
Office: NSc 319 Phone: 510-885-3879 e-mail: eric.suess@csueastbay.edu
Office Hours:
TTh 10 - 11am online on Zoom, or by appointment
Communicating:
Email is the preferred method of communication. Please do not send me emails from within Canvas. Class website will be updated weekly with class topics, homework assignment, and other useful information. Assignment grades will be provided in Canvas. Grades will be posted in Canvas.
Course Description:
Introduction to neural networks and deep learning, training a prediction model, avoiding over-fitting, tuning. Prediction and classification problems. Using R packages or Python to connect to h20, tensorflow, keras. Report Writing.
Prerequisites:
- Stat. 632
Class Website:
Required Text:
- François Chollet with Tomasz Kalinowski and J. J. Allaire, Deep Learning with R, Second Edition, Manning Publications, 2022.
Reference Texts:
- Baumer, Kaplan, Horton, Modern Data Science with R, Third Edition, CRC Press, 2024.
- Lantz, Machine Learning in R, Fouth Edition, Packt, 2023.
- François Chollet and Matthew Watson Deep Learning with Python, Third Edition, Manning Publications, 2025.
- Rick Scavetta, Deep Learning with R in Motion, Manning Publications, 2019.
Further References:
- Hadley Wickham and Garrett Grolemund, R for Data Science, 2e, O’Reilly, 2023.
- Rick Scavetta, RStudio::conf 2019, conf_tensorflow_training_day1
Technical Requirements:
Access to a modern computer and permission to install software, R, Python, RStudio. Access to Colab from your personal Google account. Access to the internet.
Material To Be Covered:
This is the fifth course in the sequence of Data Science courses offered by the Department of Statistics and Data Science for the M.S. Data Science Concentration. The Data Science courses are specifically for registered students in the M.S. Statistics program.
The sequence of courses are:
- Stat. 650 Advanced R for Data Science
- Stat. 651 Data Visualization
- Stat. 652 Statistical Learning
- Stat. 653 Statistical Natural Language Processing
- Stat. 654 Introduction to Applied Deep Learning
These courses are intended to be taken in order as they build upon each other, but you can discuss taking the courses out of order with instructor approval.
The topics of the course will follow the topics presented in the Modern Data Science with R book. The book will be used as the primary text for Statistics 650, 651, 652, 653. For each class there will be other supporting reference materials.
The main topics for Statistics 654: Applied Deep Learning
- Chapter 1. What is deep learning?
- Chapter 2. through Chapter 8.
Homework:
The material in this course can only be learned through working many problems. Most of the homework assigned during the course will be graded for content and clarity. Comments will be made and graded problems submitted that need further work should be re-submitted for a final grading.
Homework will be assigned weekly on Mondays. Homework will be due on the following Monday, which means you should complete the homework and come to class prepared to ask questions. Homework will be submitted though Canvas.
Quizzes and Exams:
Two short quizzes, one midterm will be given and the final.
Grading:
= 90% A, >= 80% B, >= 70 C, >= 60% D, <60% F
- Project 30%
- Homework 15%
- Quizzes 5%
- Midterm 25%
- Final 25%
Policy on Make-up Exams:
You are expected to take the quizzes and exams at the scheduled times. In case of genuine emergency, illness or hardship, for which you can present written documentation I may agree to arrange for a make-up exam. Make-up exams must always be arranged BEFORE the regular exam is given and always take place AFTER the regular exam. Quizzes may not be made up!
Academic Honesty:
- You are encouraged to work together on homework problems. However, each student must write up the solutions independently. Copying of solutions is not acceptable.
- You are encouraged to study together for the exams. However, each student must take the exam independently.
- Cheating will not be tolerated. Any student caught cheating will receive a reduced grade or zero for the assignment or exam in question. In addition, the student will be reported to the University for further disciplinary action.
AI Policy: Use of Generative AI is generally allowed in the course
Use of generative AI is generally welcomed in this course with appropriate citations. AI tools can be used to help brainstorm, edit, or even revise assignments. Assignments submitted without any citation of an AI tool will be considered your own original work. Assignments submitted that utilized AI tools should be appropriately referenced, otherwise it may be considered a violation of the CSUEB Academic Dishonesty Policy and may be subject to disciplinary action. (Does this policy contain a contradiction? Read the first two sentences and judge what you should cite.)
Student Learning Outcomes (SLO’s):
Student Learning Outcomes - Upon successful completion of this course students will be able to:
- Use software to learn from data.
- Critically evaluate learning models.
- Extract data from large data source to learn from data.
- Create reproducible reports.
Student Services:
To access student services offered at Cal State East Bay, click on the MyCompass to get you to your one-stop online student support hub for information on academic advising, tutoring, financial aid, the library, the health center, technology support, career counseling, campus life, equity programs, and more.
Grade Appeal and Academic Grievances:
If you wish to appeal your course grade at the end of the semester or have other academic concerns related to a course, please visit the Grade Appeals and Academic Grievances (GAAG) section of the catalog, which explains the process. URL: https://catalog.csueastbay.edu/index.php?catoid=31