- Class Times:
- Mondays, Wednesdays, and Fridays 11:00–11:50am
- First class:
- January 11, 2020
- Zoom (see eClass for link)
- James Wright (email@example.com)
- ATH 3-57
- Office hours:
- Available after class on Mondays and Fridays, and by appointment.
Introduction to modern artificial intelligence techniques, with a focus on probabilistic reasoning. Specific topics include uninformed and heuristic search, probabilistic modeling and reasoning, causal inference, deep learning, Bayesian learning, reinforcement learning, and multiagent systems. The course will emphasize the importance of appropriate choices of formal model.
- Uninformed and Heuristic Search
- Probabilistic Modeling and Reasoning
- Bayesian Learning
- Causal Inference
- Deep Learning
- Reinforcement Learning
- Multiagent Systems
After taking this survey course, you will understand the foundations of modern probabilistic artificial intelligence and how they relate to each other, in preparation for taking more advanced courses. You will understand the strengths and weaknesses of the broad families of representations in each area. You will be able to choose appropriate models in application domains, and be able to encode specific problems in those models effectively.
- Assignments: 30%
- Midterm exam: 30%
- Final exam: 40%
Assignments are to be handed in electronically via eClass by the start of lecture on the due date. Late assignments will have 20% deducted for each day that the assignment is late, up to a maximum of three days late.
Remote delivery considerations
- Synchronous lectures
- Lectures will be delivered synchronously via Zoom (see eClass for the link)
- Lectures will be recorded, to enable asynchronous access by students with connectivity issues.
- Lecture recordings will be made available on eClass until the end of the term.
- Students may turn off their camera, and are requested to turn off their audio upon joining. Students are encouraged to ask questions during lecture, either out loud or via text chat if that is more comfortable.
- Online assignments and exams
- Both assignments and exams will be submitted electronically using eClass
- An online proctoring service will not be used, but
- We will check answers for similarity, and
- Students may be called upon to explain their answers verbally in an online meeting to the instructor and/or TAs. Students who cannot explain how they arrived at their solutions may not receive credit.
Submitting the work of another person as your own constitutes plagiarism. The department is very strict about plagiarism and other academic misconduct: ALL forms of cheating are referred to the Dean’s office.
The rules for this course allow consultation collaboration. The specific rules are:
- You may discuss assignments and solutions with your classmates.
- Limit discussion to an informal verbal level. DO NOT exchange written text or source code: you can discuss assignments, but don’t look at each other’s answers or give step-by-step instructions.
- You must list all of the other students that you discussed an assignment with.
- The written part of the assignments (including source code) must be completed individually.
Students are responsible only for material that is presented in class. Slides will be made available on eClass on the day of the corresponding lecture.
Optional readings will be provided from the following texts, all of which are available online:
- David Poole and Alan Mackworth, Artificial Intelligence: Foundations of Computational Agents, 2nd edition.
- David Barber, Bayesian Reasoning and Machine Learning.
- Ian Goodfellow, Yoshua Bengio, and Aaron Courville, Deep Learning.
- Richard S. Sutton and Andrew G. Barto, Reinforcement Learning: An Introduction, 2nd edition.
- Yoav Shoham and Kevin Leyton-Brown, Multiagent Systems: Algorithmic, Game-Theoretic, and Logical Foundations.