Course syllabus
Links
Schedule Seminar Sign-up Journal Lecture Recordings
Summary
Many aspects of intelligence involve interacting with other agents. The computationalist approach to AI accordingly includes formalisms for representing models of others' mental states, mechanisms for reasoning about these formalisms, and techniques for altering them. The first portion of this course examines the role of knowledge and search in these contexts, covering topics such as collaborative problem solving; interactive dialogue; social, emotional and moral cognition; and personality. We look at a range of applications including personal assistants, human-robot interaction, training and education systems, and teams of autonomous robots. The strengths and weaknesses of this approach are brought to light in the later portions of the course via comparison to other emerging perspectives in cognitive science, philosophy of mind and AI that emphasize the dynamical, embodied and situated nature of natural intelligence.
You may also find the slides from the 5-minute "elevator pitch" of this course helpful.
Schedule
We will meet three times per week for face-to-face student/teacher learning sessions including lectures, seminars and workshops.
Day | Time | Building/Room # | |
Monday | 3-4pm | 253-101 | Maori Studies Building, room 101 |
Tuesday | 12-1pm | ALR6/421W-501 | Architecture & Planning Building (West), Room 501 |
Thursday | 12-1pm | 104-124 | Old Choral Hall, Room 124 |
Teachers
Role | Name | ext # | office location | |
Coordinator and Teacher | Dr Matthew Egbert | m.egbert@auckland.ac.nz | 87027 | 303S-491 |
Teacher | Prof Jim Warren | jim@cs.auckland.ac.nz | 86422 | 303S-483 |
Learning outcomes
This course focuses on two major movements with the broad field of AI, Interactive Cognitive Systems and Situated, Embodied, and Dynamical Systems. A successful student should be able to...
- characterise these movements within the broader context of research in artificial intelligence
- provide and explain examples of each approach
- describe the strengths and limitations of each approach
- analyze and prototype software that employs methods used in each approach
- critically evaluate literature in these and related areas
These learning outcomes relate to the target graduate profile themes in the following way
Graduate Profile Theme | Learning Outcome |
Disciplinary Knowledge and Practice | 1,2,3 |
Critical Thinking | 1,3,5 |
Solution Seeking | 3,4 |
Communication and Engagement | 1,2,3 |
Independence and Integrity | 4 |
Social and Environmental Responsibilities |
Assessment
The final exam is worth 40% of your marks. The remaining 60% is evaluated by a diverse range of coursework, including:
- 2 hands-on robotics workshops reports (10% total)
- debate report (5%)
- student-led reading-based seminar (10%)
- mid-term test (10%)
- reflective journal (5%)
- Minecraft-based AI worksheets (20%)
Requirements for Passing: A pass in both practical (the assignments) and theory (weighted mean of test and exam) is required to pass this paper. The pass thresholds are determined empirically once all marks are in.
Course summary:
Date | Details | Due |
---|---|---|