2023年12月4日

CS网课代修 | COMP9414: Artificial Intelligence Course Outline

本次澳洲网课代修主要为python人工智能相关，以下为该课程的Course Outline

Course Summary

COMP9414 is an introductory course on Artificial Intelligence covering fundamental topics such as autonomous agents, problem solving, search, logic, knowledge representation, reasoning under uncertainty, natural language processing, machine learning and neural networks. The course provides a foundation for further study in AI such as COMP4418 Knowledge Representation and Reasoning, COMP9417 Machine Learning and Data Mining, COMP9517 Computer Vision, COMP9434 Robotic Software Architecture, and COMP9444 Neural Networks and Deep Learning. Postgraduate students with more programming experience may consider enrolling in COMP9814, which is the same as the undergraduate AI course COMP3411 (offered in Term 1).

The course is taught with an orientation towards data science and with a view to practical applications of Artificial Intelligence using Python, and focuseson the use of Python toolkits for AI such as NLTK (for natural language processing) and Scikit Learn (for machine learning). Some AI applications that make use of foundational concepts will be demonstrated in lectures.

Assumed Knowledge

Students are assumed to have completed an introductory course in Python programming, such as COMP9021 Principles of Programming. Students are also assumed to have some mathematical ability, though any necessary mathematical concepts will be introduced in the course. Background knowledge of discrete mathematics would be useful.

Course Learning Outcomes

After completing this course, students will:

  1. Demonstrate understanding of the foundations of AI and fundamental AI techniques
  2. Choose appropriate AI techniques to solve given problems and implement standard AI algorithms
  3. Demonstrate practical skills in utilizing AI tookits in realistic application areas
  4. Evaluate the risks of unthinkingly applying AI in business and industry

This course contributes to the development of the following graduate capabilities:

Graduate Capability Acquired in
Scholars capable of independent and collaborative enquiry, rigorous in their analysis, critique and reflection, and able to innovate by applying their knowledge and skills to the solution of novel as well as routine problems 1–4
Entrepreneurial leaders capable of initiating and embracing innovation and change, as well as engaging and enabling others to contribute to change 1–4
Professionals capable of ethical, self-directed practice and independent lifelong learning 1–4
Global citizens who are culturally adept and capable of respecting diversity and acting in a socially just and responsible way 1–4

Teaching Strategies

The course has a 3 hour lecture and 1 hour tutorial for each of 9 topics (a week is lost due to either the Queen’s Birthday Holiday or Flexibility Week). Lectures are essential for providing conceptual foundations for the course and preparation and understanding of tutorial exercises. Tutorials are essential for understanding specific techniques, programming experience, individual feedback from tutors, and preparation for the exam.

The instruction mode of the course is fairly traditional, however the style places a greater emphasis than usual on practical applications of AI technologies and existing tools (in Python) for AI, data analytics and data science. Lectures cover more conceptual material and give an overview of the main topics and their interconnections. Lectures also provide an essential foundation to apply in tutorials and assignments. Tutorials provide students an opportunity for more in-depth analysis of particular topics and practical help on assignments. As part of tutorials, students will be given self-guided programming activities to aid in understanding core concepts. While not assessed, these lab-style exercises will help with exam preparation. Attendance is strongly encouraged to gain the maximum value from the course.

There will also be a course forum where students can post (anonymously to other students if desired) course-related questions that can be answered by staff and other students and viewed by all students. The forum will be moderated by course staff and tutors. For matters of general interest, students should use the forum in preference to e-mailing their tutor; for personal matters please e-mail the course convenor. Note that while forum questions will be endeavoured to be answered in a timely fashion, tutors work on a casual basis and cannot be guaranteed to work outside normal working hours. So before posting to the forum, please use the search functionality to check if your question has already been asked.

Teaching Rationale

COMP9414 is an introductory course that provides fundamental knowledge and skills needed for further study in Artificial Intelligence and for the application of AI in industry, particularly machine learning and data science. Where possible, concepts are augmented with practical applications of AI techniques and systems based on widely-used paradigms, tools and platforms.

The Python programming language has been chosen for several reasons: (i) this follows on from the introductory postgraduate programming course, (ii) Python is used in the machine learning and neural networks/deep learning courses, (iii) there are many practical AI toolkits built using Python, (iv) these toolkits are widely used in industry, (v) this enables the course to adopt an emphasis on data analytics/data science, a major growth area in business.

The course will use the textbook by Poole and Mackworth. This book has been chosen because: (i) it is shorter and easier to read than other AI textbooks, (ii) it adopts an approach based on logical reasoning, (iii) the book is freely available online, (iv) the book has recently been updated, and (iv) the book comes with Python implementations of standard algorithms.

Student Conduct

The Student Code of Conduct (Information, Policy) sets out what the University expects from students as members of the UNSW community. As well as the learning, teaching and research environment, the University aims to provide an environment that enables students to achieve their full potential and to provide an experience consistent with the University’s values and guiding principles. A condition of enrolment is that students inform themselves of the University’s rules and policies affecting them, and conduct themselves accordingly.

In particular, students have the responsibility to observe standards of equity and respect in dealing with every member of the University community. This applies to all activities on UNSW premises and all external activities related to study and research. This includes behaviour in person as well as behaviour on social media, for example Facebook groups set up for the purpose of discussing UNSW courses or course work. Behaviour that is considered in breach of the Student Code Policy as discriminatory, sexually inappropriate, bullying, harassing, invading another’s privacy or causing any person to fear for their personal safety is serious misconduct and can lead to severe penalties, including suspension or exclusion from UNSW.

Assessment

Item Topics Due Marks Contributes to
Assignment 1 Constraint Satisfaction Week 5/6 15% 1–3
Assignment 2 Machine Learning Week 9 25% 1–4
Final Exam All topics Exam period 60% 1–3

The two programming assignments are both auto-marked and marked manually for programming style and critical analysis. For correct auto-marking, it is the student’s responsibility to ensure that the submitted code runs on the school environment using the version of Python installed in the labs. The final exam is a 2 hour multiple-choice examination worth 60% covering the major aspects of the course. The final mark for the course is determined by adding together these component marks according to the above weighting to give a result out of 100.

Important: There is a hurdle of 40% on the exam, i.e. a mark of at least 24/60 is required in order to pass the course, otherwise a grade of UF is returned.

Late submission policy: For Assignments 1 and 2, late submissions are subject to the penalty that the maximum mark you can obtain reduces by 20% of the value of the assignment per day or part-day late, for up to 3 calendar days, after which a mark of 0 is given.

Course Schedule

There are 9 topics in the course, with a lecture and tutorial on each topic. There will be one week with no lecture, which will be either Week 2 or 3 (because of the Queen’s Birthday Holiday) or Week 6 (Flexibility Week), depending on whether lectures are scheduled on Monday or on another day.

Topic Lecture Tutorial Assignment
1 Artificial Intelligence and Agents Agents
2 Problem Solving and Search Search
3 Constraint Satisfaction Problems Constraint Satisfaction/Planning
4 Logic and Knowledge Representation Propositional Logic
5 Reasoning with Uncertainty Uncertainty Assignment 1
6 Machine Learning Machine Learning
7 Natural Language Processing Language Processing
8 Knowledge Based Systems First Order Logic Assignment 2
9 Neural Networks and Reinforcement Learning Neural Networks/Reinforcement Learning

Resources for Students

The textbook for this course is freely available online. We will also draw on material from the COMP3411 textbook (Russell & Norvig), which provides much more detail than is needed in this course, but covers some topics especially well, particularly search. It is recommended as a comprehensive reference.

The course draws on the use of two Python AI packages, namely NLTK (Natural Language Tookit) and Scikit Learn. Scikit Learn will be used in COMP9417 Machine Learning and Data Mining.