CSci 356 Course Syllabus
Linux Lab (Woods 136)
MW 2:00 pm
|
Prof. Stephen P. Carl
|
Office: |
Woods 133 |
Availability: |
Tues
2:00-4:00, Wed 3:15-4:00 after class
Also by appointment, using Calendly
|
E-mail: |
scarl @T sewanee D0T edu |
Phone: |
931.598.1305 |
|
Course Objectives: the student
will understand
- knowledge representations for AI systems
- various forms of state-space search, including DFS, BFS, and heuristic
search
- biology-inspired methods of machine learning such as neural
networks and genetic algorithms
- layering of supervised and unsupervised learning methods to create deep
learning networks
Textbooks.
Course website: https://scarl.sewanee.edu/CS356/
The course schedule, lecture notes, assignments, and this syllabus are
all posted on the course website. Video lectures (if needed) and other
material will be posted on BrightSpace.
Grading
Workload
|
Points
|
6 Tutorial Assignments (5 points
each) |
30 points
|
2 Programming
Projects (10 points each) |
20 points
|
Research Summary |
8 points
|
AI Paper + Presentation |
8+4
points
|
Midterm Exam
|
15 points
|
Final Exam |
15 points |
Policies - Quick Look
- Attendance is an important
factor in doing well in the class as there will be the occasion in-class
exercise; the student is responsible for making up any work missed due
to absence.
The Dean's Office may be notified after three
unexcused absences.
- Tutorial assignments
are used to gain experience with a specific AI programming technique.
Students are to work independently unless group work is
specifically indicated. Assignments will typically be turned in using BrightSpace,
though occasionally I may require a printed copy of program files.
- Late assignments are penalized 10% for each day late,
but every student has 3 grace days for the semester, covering
things like schedule crunches, road games, illnesses. Save these as
long as possible. No assignment more than 4 days late will
be accepted for grading.
- Lecture summaries: to support the intellectual life of the
university, I encourage you to attend at least two
lectures/talks outside of the formal classroom experience and submit a
1-2 page summary of the ideas
presented, worth up to 4 points of extra credit.
- The Honor Code applies to all exams and assignments. Plagiarism
(defined below) applies to any medium submitted, electronic or
otherwise. See below for more information.
- Course Administration:
- The first part of this course will mainly consist of lectures and a
few in-class exercises to become familiar with the concepts. The second
part will follow more of a workshop model: short lectures (and possibly
an occasional video lecture) followed by at least one in-class
exercise or programming lab per week. One of the main
ways you'll learn this material is by getting knee-deep into it
yourselves.
- Assignments
- Assignments will typically involve interacting with an existing AI
program written in Python, and modifying it in various ways. Later in
the semester we'll be working almost exclusively with Jupyter
Notebooks, experimenting with Deep Learning programs using the Keras
API.
- Avoiding plagiarism: Students may discuss ideas for solving an
assignment among themselves: concepts and design issues, how to use
software tools, how to fix compilation errors. However, each assignment
must be your own work unless
collaboration is specifically allowed. Turning
in any portion of work written by another is an Honor Code violation
and grounds for disciplinary action as allowed by University policy.
You should not copy a file, supply a copy of a file, coach another
student in writing code line by line, or look at another's code. A
good rule of thumb: explaining how to fix a coding problem in
English is fine, sitting down and fixing someone else's code is not. See
below for the official Univerity policy.
- For lecture summaries you'll
attend and report on up to two lectures outside class, such as our
department's Ebey
lecture or lectures sponsored by other departments. You might
also find useful technical presentations online, such as the
Heidelberg Laureate series, the Strange Loop conference channel on
YouTube, and other computing-related meetings for researchers or
software developers.
- A
BrightSpace Explainer
- Since the pandemic, BrightSpace
(Sewanee's Learning Management System) has been the entry point into
our courses. I've posted a link to the course website, lecture
videos, and space for submitting assignments.
- Scheduling
Office Hours
- Students are welcome to meet me in
my office during the office hours posted above; otherwise, schedule a
time to meetup in my office or by Zoom. I use the
Calendly app for scheduling; see the video
on their website for a quick introduction to this service.
- Honor Code Statement
- When you matriculated, you agreed to follow the Honor Code: to do your
own honest work and not to cheat in any form. All forms of cheating,
including plagiarism, are violations of the Honor Code and will be
treated as such. As per the Honor Code, plagiarism is defined as
"[copying or imitating] the language and thoughts of others and
[passing] the result off as an original work." Using the language or
ideas of others without proper citation is considered academic
dishonesty (cheating), and "others" includes responses from artificial
intelligence processing programs (for example, ChatGPT). As well, using
AI to complete assignments without the express and clear approval of
your instructor is also a violation (for receiving unauthorized
assistance). If you ever have a question about an assignment or need
additional help, please ask for the instructor for assistance rather
than jeopardize your academic career.
- ADA
Statement.
- The University of the South is committed to fostering respect for the
diversity of the University community and the individual rights of each
member of that community. In this spirit, and in accordance with the
provisions of Section 504 of the Rehabilitation Act of 1973 and the
Americans with Disabilities Act (ADA), the University seeks to provide
students with disabilities with the reasonable accommodations needed to
ensure equitable access to the programs and activities of the
University.
- Any student with a documented disability needing academic adjustments
is requested to speak with Student Accessibility Services (SAS) as early
in the semester as possible, though students may reach out at any point
during the academic year. If approved for accommodations, the student
has the responsibility to present their instructors with a copy of the
official letter of academic accommodations. Please note: Accommodation
letters should be dated for the current term; accommodations will not be
provided without a current accommodation letter; and accommodations
cannot be applied retroactively.
- SAS is located in the Office of the Dean of Students (931.598.1229).
Additional information about accommodations can be found on the
Student Accessibility Services website.