Artificial Intelligence
COMP9414
Lecturers
• Dr. Francisco Cruz (LiC – Lecturer in Charge)
• https://www.unsw.edu.au/staff/francisco-cruz-naranjo
• Dr. Armin Chitizadeh
• https://www.unsw.edu.au/staff/armin-chitizadeh
Course admin
• Maryam Hashemi
• https://maryamhashemi1995.github.io/index3.html
Tutors
• Kane Walter kane.walter@unsw.edu.au
• Zahra Donyavi z.donyavi@unsw.edu.au
• Madeleine Nouri m.nouri@student.unsw.edu.au
• Carson Liu carson.liu@unsw.edu.au
• Stefano Mezza s.mezza@unsw.edu.au
• Lina Phaijit l.phaijit@unsw.edu.au
• Jingying Gao jingying.gao@unsw.edu.au
• Adam Stucci a.stucci@unsw.edu.au
• Raktim Kumar Mondol r.mondol@unsw.edu.au
• Siti Mariyah
• Session will be BYOD. Alternatively you can borrow a laptop. See
https://taggi.cse.unsw.edu.au/FAQ/Borrow_A_Laptop/
Artificial Intelligence (AI)
• What is intelligence?
Artificial Intelligence (AI)
• What is intelligence?
• it can be described as the ability to perceive or infer information, and to
retain it as knowledge to be applied towards adaptive behaviours within an
environment or context [Wikipedia].
Artificial Intelligence (AI)
• What is intelligence?
• it can be described as the ability to perceive or infer information, and to
retain it as knowledge to be applied towards adaptive behaviours within an
environment or context [Wikipedia].
• What is artificial intelligence?
Artificial Intelligence (AI)
• What is intelligence?
• it can be described as the ability to perceive or infer information, and to
retain it as knowledge to be applied towards adaptive behaviours within an
environment or context [Wikipedia].
• What is artificial intelligence?
• Artificial intelligence (AI) is intelligence demonstrated by machines, as
opposed to intelligence displayed by humans or by other animals [Wikipedia].
Artificial Intelligence (AI)
• What is intelligence?
• it can be described as the ability to perceive or infer information, and to
retain it as knowledge to be applied towards adaptive behaviours within an
environment or context [Wikipedia].
• What is artificial intelligence?
• Artificial intelligence (AI) is intelligence demonstrated by machines, as
opposed to intelligence displayed by humans or by other animals [Wikipedia].
• Can you give some examples?
AI is not ML is not ANN
AI is not Python
Current (mis)understanding of AI
How are currently perceived AI-based systems?
Current (mis)understanding of AI
But in reality, it’s still an open problem
Course Plan
• Introduce AI concepts through intelligent agents
• Start with very simple reactive agents
• Progress by adding more capabilities
• End with agents that learn, reason and
communicate
Week 1
1 Introduction
1.1 History of AI
1.2 Agents
1.2.1 Reactive agent
1.2.2 Model-based agent
1.2.3 Planning agent
1.2.4 Utility-based agent
1.2.5 Game-playing agent
1.2.6 Learning agent
Week 2
2 Search
2.1 Uninformed search
2.2 Informed search
2.3 Informed vs uninformed
Week 3
3 Rewards instead of goals
3.1 Elements of reinforcement learning
3.2 Exploration vs exploitation
3.3 The agent-environment interface
3.4 Values functions
3.5 Temporal-difference prediction
Week 4
4 Neural Networks
4.1 Neurons - biological and artificial
4.2 Single-layer perceptron
4.3 Linear separability
4.4 Multi-layer networks
4.5 Backpropagation
4.6 Neural engineering methodology
Week 5
5 Metaheuristics
5.1 Asymptotic complexity
5.2 Classes of problems
5.3 Linear programming
5.4 Search space
5.5 Metaheuristics with and without memory
5.6 Genetic algorithms
Week 6
Recap and consultation
Week 7
7 Robot vision
7.1 Image processing
7.2 Scene analysis
7.3 Cognitive vision
Week 8
8 Language processing
8.1 Chomsky’s hierarchy
8.2 Formal languages
8.2.1 Regular expressions
8.2.2 Grammars
8.3 Natural languages
8.2.1 Language and linguistic landscape
8.2.2 Conversational agents
Week 9
9 Reasoning with uncertain information
9.1 Probability and probabilistic inference
9.2 Bayes nets
9.3 Learning Bayes nets
Week 10
10 Knowledge representation
10.1 Feature-based vs iconic representations
10.2 Logic
10.3 Learning rules
Course Plan
Related Course
• COMP3431 Robot Software Architectures
• COMP4418 Knowledge Representation and Reasoning
• COMP9417 Machine Learning and Data Mining
• COMP9444 Neural Networks and Deep Learning
• COMP9491 Applied Artificial Intelligence
• COMP9517 Machine Vision
Timetable
• Lecture:
• Tue 16:00 – 18:00
(Physics Theatre)
• Consultation time:
• Tue 11:00 – 12:00
(J17 Lv 5 Rm 510J)
• Tutorials:
Important dates
• First lecture: Tuesday 29th May 2023
• Last lecture: Tuesday 1st August 2023
• Assignment 1 open: Friday 9th June 2023
• Assignment 1 deadline: Friday 30th June 2023
• Assignment 2 open: Monday 10th July 2023
• Assignment 2 deadline: Monday 31st July 2023
• Exam: Exams period
Assessment
• Assessment will consist of:
• Assignment 1: results (10%), discussion (10%).
• Assignment 2: results (10%), discussion (10%).
• Final exam (60%).
• To pass, you must score:
• At least 16/40 for the assignments.
• At least 24/60 for the exam.
• A combined mark of at least 50/100.
Student Conduct
• Assignments can be done individually or in couples.
• There will be no difference in the assessment.
• Both students must participate in the discussion.
• Late deliveries will be accepted subject to 5% discount per day from
the results (including weekends and public holidays), for up to 5 days,
after which mark is 0.
• It’s students' responsibility to have code discussions with tutors in
time.
• Plagiarism is academic misconduct.
Contact
• The first contact should be the forums.
• Additionally, a consultation time is scheduled every week.
• In special circumstances you could also email to the lecturing team
(cs9414@cse.unsw.edu.au)
Texts & References
• Poole, D.L. & Mackworth, A. Artificial Intelligence: Foundations of
Computational Agents. Second Edition. Cambridge University Press,
Cambridge, 2017.
• Russell, S.J. & Norvig, P. Artificial Intelligence: A Modern Approach.
Fourth Edition, Pearson Education, Hoboken, NJ, 2021.
• Sutton, R. & Barto, A. Reinforcement Learning: An Introduction. MIT
press. 2018.
• Jurafsky, D. & Martin, J. H. Speech and Language Processing. Stanford.
2023.
Questions?