CS3600 Intro to Artificial Intelligence
CS3600 Intro to Artificial Intelligence
Contact: hyeokhyen@gatech.edu
Office hours: Mon (CoC commons) 02:00 pm - 04:00 pm
Wed (TSRB226) 02:00 pm - 04:00 pm
and by appointment
Contact: shiremath9@gatech.edu
Office hours: Thursday (CoC commons) 3-5:00 pm
Contact: jhaver.shagun@gatech.edu
Office hours: Thursday (CoC commons) 4-6:00 pm
CS3600 is a 3-credit introductory course for undergraduates which is a broad overview of Artificial Interlligence (AI) by employing agent-based approach. Over the course of the semester, students will learn about the methods and tools to design and build an intelligent agent that can understand, reason, learn about the world, and act rationally in the given environment. It is a programming intensive class, focusing on the details of implementation with Python.
To provide a broad survey of algorithms and approaches in AI.
To develop a deeper understanding of several major algorithms in AI.
To develop the design and programming skills that will help you to build intelligent artifacts.
To develop the fundamental skills and backgrounds necessary to pursue research in Intelligent Systems (IS). The fields of research in IS includes planning, knowledge representation, machine learning, vision, robotics, and etc.
To provide a useful foundations for a number of courses involving Intelligent Systems (IS), which are Machine Learning (CS4641), Knowledge-Based AI (CS4634) , Computer Vision (CS4495), Robotics and Perception (CS4632) , Natural Language Understanding (CS4650), Game AI (CS4731).
Official prerequisite is CS1332 Data structures and Algorithms.
Or at least, capability of reading pseudocodes is required.
Also, familiarity with probability theory, linear algebra, and calculus is welcomed.
Textbook: Artificial Intelligence: A Modern Approach by Stuart Russell and Peter Norvig, Third edition. 2010. ISBN: 978‑0136042594. (the blue book)
The course will mainly follow a number of chapters in the textbook closely. Lectures are meant to summarize the readings and stress the important points. Covering all chapters are not possible with given time. But highly recommend to read the uncovered chapters yourself, if you intend to pursue AI research career.
The occasional supplemental readings will be provided to you.
Web. We will use Canvas for project submission and important announcements. And Piazza will be used for manging the course and discussing general questions.
Homeworks: 0% (Keeping up with it helps you do well on the exams)
Projects: 60% (4 projects and 15% each)
Midterm: 20% (You are allowed one 8.5x11in sheet of notes, front and back)
Final: 20% (You are allowed one 8.5x11in sheet of notes, front and back)
Participation: 0% (But, it will be referred to determine whether your grade can be lifted in case you are right on the edge of two grades.)
All the assignments are individual work only. We do check similarity of your code with the solutions on the Internet which is regularly updated (line by line). Hence, we suggest you not even search for solutions because once you find them, it is difficult to ignore the rest of the solution. Any plagiarism is the violations of the honor code.
You may discuss with other students in the class about the assignments. We suggest interacting with a whiteboard or a piece of paper, and erasing the discussion. We highly recommend not taking picture of the board or the paper used. This will prevent you from copying each others' code directly when discussing on the computer. Discovering copies of code between students will lead all students involved to be considered as violating the honor code.
Make sure to store your solutions in a secure repository or your local machine and do not post on online public repositories, such as Bitbucket or Girhub. This is to prevent any possibility of other students referring or copying your solutions.
If you have any difficulties with the assignments including the deadline conflict with other courses' assignments or personal emergencies, contact me or the TAs via email or in person. We will try to find ways to assist you, such as deadline extensions, extra office hours, and etc.
Disclaimer: The following is the tentative schedule for Summer 2019 semester. Since we only have fewer weeks than Spring and Fall, this can change on demands. (Check your e-mail & Canvas). Also, please read chapters from the textbook for each week given below. This will help you better digest the course contents, and hence, help me keep the pace of the lecture. Any active participation in class will be recorded!
Week 1 (May 14th & 16th): Introductions and Logistics & Search. Reading: Ch 1-3.
Week 2 (May 21th & 23th): Search practice & Markov Decision Processes. Reading: Ch 2-3. [Project 0 due]
Week 3 (May 28th & 30th): Markov Decision Processes. Reading: Ch 17.1-17.3
Week 4 (June 4th & 6th): Q-learning. Reading: Ch 21.3.2.[Project 1 due (Jun 7th)]
Week 5 (June 11th & 13th): Probability review & Bayes nets. Reading: Ch 13 & Ch 14.
Week 6 (June 18th & 20th): Midterm review (18th) & exam (20th) during normal class period
Week 7 (June 25th & 27th): Filtering Reading: Ch 14. [Project 2 due (June 28th)]
Week 8 (July 2nd): Optimization. Reading: Ch 4.
Week 9 (July 9th & 11th): Machine Learning & Decision Trees. Reading: Ch 18. [Project 3 due (July 12th)]
Week 10 (July 16th & 18th): Neural Networks & POMDP Reading: Ch 18 & Ch 17.4
Week 11 (July 23rd): Course summary and final review. [Project 4 due by 11:59pm EDT on Sunday July25th.]
Final exam (July 26th, Friday, 8:00 AM - 10:50AM)
All projects submitted through Canvas (each due dates are specified). Submission by E-mail is not accepted. One hour late submission is accepted with 10% penalty.
Project 0: Tutorial project for using Python and autograder (The ungraded project.)
Project 1: Solve for various searching tasks
Project 2: Use MDPs and Reinforcement Learning for solving various tasks in an environment with stochastic action sequence.
Project 3: Inference and filtering via Bayesian Networks
Project 4: Machine Learning task with Decision Trees.