Logistics
Objectives
- understand the significance of commonsense in modern AI
- master basic knowledge of cognitive science for modern AI
- derive statistical computational models of commonsense to solve challenging AI problems
- code and build complex system to model certain aspect(s) of commonsense
Prerequisites
- basic knowledge of statistics, e.g., Bayes rule
- comfortable with programming in Linux using modern deep learning libraries
- excellent English skills for reading literature, presentation, and writing homework and reports
- familiar with LateX for writing homework and reports
Grading
- strict rule: no late submission will be accepted after the due date
- paper presentation (in English): 20%
- essay (in English): 3% $\times$ 10 topics = 30%
- project presentation (in English): 20%
- project report (in English and LaTeX): 30%
Essay (30%)
- team: 1 student; work alone
- duration: 1 week after each lecture
- topic: choose one from each lecture, 10 in total
- grading: 3 points per essay
- delivery: essay (2 pages minimal) in LaTeX, written in English
Paper Presentation (20%)
- team: 2-4 students
- duration: up to the presentation session
- topic: choose papers listed in one of the 10 topics and additional papers at your own choice
- note: no need to cover all the listed papers, but need to convey a coherent message during the presentation
- delivery: presentation, in English
Project (50%)
- team: 2-4 students, usually the same team as for the paper presentation
- duration: the entire semester
- topic: choose one listed on the project page (see also the instructions); topics are exclusive among teams; reserve online
- delivery:
- presentation in English
- technical report in top conference quality that includes
- insights of the problem
- related work / literature review
- major components of the systems of algorithms
- technical details, e.g., learning, training, parameters
- experimental results
- publicly available code, e.g., on GitHub
Plagiarism
- please refer to the plagiarism page for details.
Errata/Typo
- please contact the instructor or TA
Acknowledgement
This course is made possible by a group of collaborators and students
my advisors,
my long-term collaborators,
- Prof. Tao Gao (UCLA)
- Prof. Federico Rossano (UCSD)
- Prof. Hongjing Lu (UCLA)
the most awesome peers in the world (in alphabetical order),
- Dr. Lifeng Fan (BIGAI)
- Dr. Siyuan Huang (BIGAI)
- Dr. Baoxiong Jia (BIGAI)
- Dr. Hangxin Liu (BIGAI)
- Dr. Tengyu Liu (BIGAI)
- Dr. Siyuan Qi (BIGAI)
- Dr. Luyao Yuan (Meta)
- Dr. Zeyu Zhang (BIGAI)
- Dr. Zilong Zheng (BIGAI)
- Dr. Chi Zhang (BIGAI)
the smartest students/postdocs in the world (in alphabetical order),
- Dr. Bo Dai (PKU, BIGAI)
- Guangyuan Jiang (PKU, MIT)
- Shiqian Li (PKU, BIGAI)
- Yuyang Li (THU, BIGAI)
- Shuwen Qiu (UCLA)
- Liangru Xiang (THU, BIGAI)
- Chao Xu (UCLA)
- Dr. Fangwei Zhong (PKU, BIGAI)
[22530005/04804023] Syllabus
Introduction
- Week 01/Lecture 01: Introduction & Logistics
Physical Commonsense
- Week 02/Lecture 02: Affordance and Functionality
- Week 03/Lecture 03: Intuitive Physics
- Week 04/Lecture 04: Causality
- Week 05/Lecture 05: Tool, Mirroring, and Imitation
- Week 06/Paper Presentation 1: Physical Commonsense
Social Commonsense
- Week 07/Lecture 06: Communication
- Week 08/Lecture 07: Intentionality
- Week 09/Lecture 08: Animacy and Theory of Mind (ToM)
- Week 10/Paper Presentation 2: Social Commonsense
Advanced Topics
- Week 11/Lecture 09: Abstract Reasoning
- Week 12/Lecture 10: Utility
- Week 13/Lecture 11: Explainable AI and Teaming
- Week 14/Paper Presentation 3: Advanced Topics
Project Presentations
- Week 15/Project Presentation 1
- Week 16/Project Presentation 2