Cal Poly SLO CSC 570-W18 Computational Intelligence Prof. Franz J. Kurfess
CSC 570 Computational Intelligence Syllabus
Dr. Franz J. Kurfess, Cal Poly Computer Science & Software Engineering Department (http://www.csc.calpoly.edu/~fkurfess/)
My office hours are tentatively scheduled as follows:
You can find up-to-date information on my online calendar.
According to the catalog, this is a seminar-style class, and does not have an official lab period. This class will include a team project, and part of the class time will be used for project work.
The Cal Poly Catalog 2013-15 describes the course as follows:
Prerequisite: Graduate standing and evidence of satisfactory preparation in computer science.
Directed group study of selected topics for graduate students. Topics will normally consist of continuations of those in CSC 520, CSC 530, CSC 540, CSC 550, CSC 560 and CSC 580, and other topics as needed. Class Schedule will list topic selected. Topic credit limited to 12 units. 2 to 4 seminars.
From related classes (primarily CSC 480), students should be familiar with the following topics:
If necessary, these topics can be reviewed by looking at the lecture notes of the CSC 480-F17 course.
The course will cover recent developments in the area of Artificial Intelligence, with particular emphasis on advances triggered by the use of approaches characterized as “deep learning”.
The goal of this course is tan in-depth examination of recent approaches to solve problems in the area of Artificial Intelligence with methods based on neural networks and often utilizing massive data sets in combination with significant computing power. Students will select specific areas of interest, and identify relevant work performed in those areas. This work may include conventional research publications, but also software and data sets made available by other parties.
Through this course, students are expected to:
The choice of topics will depend significantly on the areas of interest selected by the students. Most of the class time will be used for student presentations including discussions based on those presentations, and project-related activities.
Due to the nature of the topics to be discussed, and the incorporation of very recent developments in the field, this class will not use a traditional textbook. As background material, in particular for students who have not taken the respective courses, I suggest to consult the respective sections of the CSC 480 Artificial Intelligence and CSC 580 Intelligent Agents courses.
Here is one book on the topic that is fairly up to date:
Since one of the focus areas of the course will be deep learning, here are some resources:
And here are a few more pointers to specific tools like TensorFlow:
The main work in this class consists of a team project and a research activity.
Student teams will have several project topics to choose from, with an emphasis on mobile devices. Some of the projects can be done in collaboration with outside partners. The project work should focus on Artificial Intelligence, and I recommend to coordinate the research activities of the team members with the project topic. The teams are expected to design and implement a system that uses AI methods to solve a practical problem. This can range from a proof of concept for a novel idea, the application of proven approaches to a new domain or problem, or improvements to an existing system. Usually the project outcomes will be shown in a display around the mid-quarter point and one at the end of the quarter. Details will be discussed during the first or second week of the quarter.
This course is an advanced graduate course on Computational Intelligence, and students are expected to investigate a topic related to important principles and recent work in the field. Traditionally, the results of such research work are delivered in the form of a presentation and research paper. In this class, we will examine alternative approaches to present the outcomes of the research conducted as class work. This can be in the form of an entry to Wikipedia or a similar Web site, a series of blog entries, a video, a podcast, or of course a traditional paper. The activity can be conducted individually, in small groups, or by the same team that works on the project. If multiple students collaborate, the contributions of each student must be clearly identified. For presentations, each group member is expected to do a part of the presentation.
It is subject to the following expectations:
This class will rely on interactive classroom activities, such as participation in group discussions, presentation of ideas and results (from the research activities or team projects), providing written summary materials (as web files via Piazza or Moodle), etc.
Success in this class depends on regular attendance, preparation of assigned readings and homework exercises, as well as a level of professionalism in the class presentations and displays. Peer evaluations may be included as part of the grade.I will use the following allocation of scores for the calculation of the grades.
I reserve the right to change the formula used. Please note that the project consists of several parts which will be evaluated separately. The project will be done in teams, and the performance of the team as a whole will be graded unless there is a clear disparity in the contribution of the individual team members. Should this be the case, I might ask for additional documentation like work sheets, email messages, or draft copies of documentation to evaluate individual contributions. For the team grades, feedback through peer evaluations will also be considered (although I will not use it directly in the calculation of the score).
The official final exam dates and times are listed on the schedule. The following activities may be scheduled for this final exam time; details will be discussed in class:
Students are expected to attend all lecture and lab sessions for the class. Except for unforeseeable reasons like illness or accidents, I expect advance notice for anticipated absences and delays in submission of class work.
To maintain uniformity across the student population, I am following university guidelines and will consider the following “excusable” reasons for allowing students to make up missed work and absences:
Much of the graded work in this class depends strongly on presentations and documentation material. Once a team or individual has committed to a date for the presentation, extensions or changes in the dates will only be permitted for the reasons listed above. Such changes may also have to be coordinated with the project contacts at the outside partner.
It is University policy to provide, on a flexible and individualized basis, reasonable accommodations to students who have disabilities that may affect their ability to participate in course activities or to meet course requirements. Students with disabilities are encouraged to contact their instructor to discuss their individual needs for accommodations. If you have a disability for which you are or may be requesting an accommodation, you are encouraged to contact both your instructor and the Disability Resource Center, Building 124, Room 119, at (805) 756-1395, as early as possible in the term.
The expectations below are based on Cal Poly’s Code of Student Conduct.
Academic dishonesty, in particular plagiarism, can be a serious offense. Any instances of cheating or plagiarism may be reported to the department chair and the Office of Student Rights & Responsibilities (OSRR). The Cal Poly rules and policies are listed in the Cal Poly catalog as well as at the OSRR web site. If the rules are unclear or you are unsure of how they apply, ask the instructor beforehand.
For programming assignments, we may use programs or services like Moss to compare assignments within a section, across all current sections of this class, and with old assignments. While such programs are not perfect, they detect suspicious similarities even after replacement of variable names and other identifiers. In general, the use of program libraries is acceptable, but not if they provide functionality whose implementation is the purpose of the lab or assignment. If you use libraries you need to indicate this in the documentation.
Turning in work is presumed to be a claim of authorship unless explicitly stated otherwise. For work created by multiple persons (such as team projects or group presentations), I may ask for documentation on who is responsible for which parts or aspects of the work.