Objective

This course offers a gentle introduction to the concepts and methodologies in Artificial Intelligence from both theoretical and practical perspectives. This includes designing intelligent agents through techniques such as state-space search methods, knowledge representation and inference, and Machine Learning. By the end of this course, students are expected to have the ability to develop intelligent solutions for semi-real world problems using the appropriate methods discussed in the course. Furthermore, they are expected to gain knowledge and experience in analyzing the inner workings of these methods and customizing them for specific problems.

Contacting Instruction Team

Please contact us through the Telegram channel at https://t.me/ai_spring_2024 and for necessary cases, email the respective heads of teaching assistance.(Ensure that you use a proper subject for your email and include your name and student ID. It is strongly encouraged to use the @Sharif webmail service to ensure easy identification by the instructional team.)
- arash.marioriyad98@sharif.edu
- s.emad.emamjomeh@gmail.com
- mojtabanafez96@gmail.com
- ilyamahrooghimath@gmail.com
Note that you can always directly contact the instructor via his email address.

Previous Websites
Content
Prerequisites

Knowledge and experience of a general-purpose programming language (Python), Data Structures and Algorithms, and ideally Probability and Statistics (Random Variables, (Joint) Probability Mass Functions, Conditional Probability, Independence, Bayes Theorem) and Linear Algebra. It is highly recommended to have the Engineering Probability and Linear Algebra courses passed before taking this course.

Classes

The class is scheduled for Sundays and Tuesdays from 13:30 to 15:00. By default, classes will be held in person; however, under special circumstances, the class might be held virtually with prior announcements. The virtual class link is https://vc.sharif.edu/ch/rohban.

Grading

Please note that grades will be calculated out of $22.0$, with the respective grades for each section provided below:

  • Homework Assignments: $7.5 + 0.5$ points.
    • Five Series, each $1.5$ points and $0.5$ bonus point in total.
  • Presentation: $1.0$ point.
  • Quiz: $1.5$ points.
  • Midterm Exam: $4.5$ points.
  • Final Exam: $5.5$ points.
  • Project: $+1.5$ point.
    • The project is optional but offers the advantage of bonus points.

Notably, achieving at least 40% of exam total grades (Midterm+Final) is required to pass the course.

Exams

Students' learning will be assessed through two quizzes, a midterm, and a final exam. These exams will be held on the following dates, and participation is mandatory for all students.

  • 1$^{\text{st}}$ Quiz : 1403/01/26 13:30 PM.
  • 2$^{\text{st}}$ Quiz : 1403/03/06 13:30 PM.
  • Midterm Exam: 1403/02/13 13:30 PM.
  • Final Exam: 1403/04/03 9:00 AM (as scheduled in EDU).
Homeworks

Five series of homework assignments will be distributed in this course. Assignments will be released on Saturday midnight every other week. Students will have about 20 days to submit their answers on Quera. Regarding the late submission policy, you can submit answers with total delay of 10 days with no penalty. Any additional delays will lead to a 0.5% reduction in the assignment grade for every hour of delay. This is illustrated in the figure below, where ℎ represents the amount of delay in hours

Delay Coefficient graph
To prioritize fairness and consistency for all students, a request for a deadline extension must be emailed to the instructor and head TAs at least 72 hours before the deadline.

Collaboration, Academic Honesty, and Plagiarism

There will be a zero-tolerance policy for cheating or copying homework assignments. The first time you are caught, we will refer you to the committee, and you will fail the course. Please refer to the Education Committee's statement on homework etiquette.

Course Pages
  • https://sut-ai.github.io/: This is the main page of our course and all exercises, slides and course resources will be placed in it.
  • Quera, https://quera.org/course/add_to_course/course/16535/: Delivery of exercises and announcements will be done entirely through Quera. Make sure you enter an email on Quera that you check regularly. Your questions from the exercises will also be answered in Quera.
Schedule

1

Date:

Sun Bahman 15

Description:

Introduction & Intelligent Agents

Course Materials: Events: Deadlines:

2

Date:

Tue Bahman 17

Description:

Uninformed Search

Course Materials: Events: Deadlines:

3

Date:

Tue Bahman 24

Description:

Uninformed Search

Course Materials: Events: Deadlines:

4

Date:

Sun Bahman 29

Description:

Informed Search

Course Materials: Events: Deadlines:

5

Date:

Tue Esfand 1

Description:

Local Search

Course Materials: Events: Deadlines:
Date:

Sat Esfand 5

Description: Course Materials: Events:

HW1 Out [handout]

Deadlines:

6

Date:

Tue Esfand 8

Description:

Local Search

Course Materials: Events: Deadlines:

7

Date:

Sun Esfand 13

Description:

Search in Continuous spaces

Course Materials: Events: Deadlines:

8

Date:

Tue Esfand 15

Description:

Constraint Satisfaction Problems I

Course Materials: Events: Deadlines:

9

Date:

Sun Esfand 20

Description:

Constraint Satisfaction Problems II

Course Materials: Events: Deadlines:

10

Date:

Tue Esfand 22

Description:

Adversarial Search

Course Materials: Events: Deadlines:
Date:

Fri Esfand 25

Description: Course Materials: Events: Deadlines:

HW1 Deadline

Date:

Sat Esfand 26

Description: Course Materials: Events:

HW2 Out [handout]

Deadlines:

11

Date:

Tue Farvardin 14

Description:

Uncertainty & Inference

Course Materials: Events: Deadlines:
Date:

Fri Farvardin 17

Description: Course Materials: Events: Deadlines:

HW2 Deadline

Date:

Sat Farvardin 18

Description: Course Materials: Events:

HW3 Out [handout]

Deadlines:

12

Date:

Sun Farvardin 19

Description:

Bayesian Networks: Representation

Course Materials: Events: Deadlines:

13

Date:

Tue Farvardin 21

Description:

Inference in Bayesian Networks: Exact

Course Materials: Events: Deadlines:

14

Date:

Sun Farvardin 26

Description:

Inference in Bayesian Networks: Approximate

Course Materials:

Sessions 1-8

Events:

Quiz 1
[handout]
[solution]

Deadlines:

15

Date:

Tue Farvardin 28

Description:

Temporal Probability Models: Markov Chains & HMMs

Course Materials: Events: Deadlines:

16

Date:

Sun Ordibehesht 2

Description:

Temporal Probability Models: Particle Filters

Course Materials: Events: Deadlines:

17

Date:

Tue Ordibehesht 4

Description:

Learning in Bayesian Networks & Naive Bayes

Course Materials: Events: Deadlines:
Date:

Fri Ordibehesht 7

Description: Course Materials: Events: Deadlines:

HW3 Deadline

Date:

Sat Ordibehesht 8

Description: Course Materials: Events: Deadlines:

18

Date:

Sun Ordibehesht 9

Description:

Decision Tree

Course Materials: Events: Deadlines:

19

Date:

Tue Ordibehesht 11

Description:

Concepts of Machine Learning

Course Materials: Events: Deadlines:
Date:

Thu Ordibehesht 13

Description:

Midterm Exam
[handout]
[solution]

Course Materials:

Sessions 1-13

Events: Deadlines:
Date:

Sat Ordibehesht 15

Description: Course Materials: Events:

HW4 Out [handout]

Deadlines:

20

Date:

Sun Ordibehesht 16

Description:

Regression & Optimization

Course Materials: Events: Deadlines:

21

Date:

Tue Ordibehesht 18

Description:

Perceptron

Course Materials: Events: Deadlines:

22

Date:

Sun Ordibehesht 23

Description:

Neural Networks I

Course Materials: Events: Deadlines:

23

Date:

Tue Ordibehesht 25

Description:

Neural Networks II

Course Materials: Events: Deadlines:

24

Date:

Sun Ordibehesht 30

Description:

Markov Decision Process

Course Materials: Events: Deadlines:

25

Date:

Tue Khordad 1

Description:

Markov Decision Process: Value Iteration & Policy Iteration

Course Materials: Events: Deadlines:
Date:

Fri Khordad 4

Description: Course Materials: Events: Deadlines:

HW4 Deadline

Date:

Sat Khordad 5

Description: Course Materials: Events:

HW5 Release

Deadlines:

26

Date:

Sun Khordad 6

Description:

Reinforcement Learning: Passive & Active

Course Materials: Events: Deadlines:

27

Date:

Tue Khordad 8

Description:

Reinforcement Learning: Approximate

Course Materials: Events: Deadlines:
Date:

Wed Khordad 9

Description: Course Materials:

Sessions 14-23

Events:

Quiz 2

Deadlines:
Date:

Fri Khordad 25

Description: Course Materials: Events: Deadlines:

HW5 Deadline

Date:

Sun Tir 3

Description:

Final Exam

Course Materials:

Sessions 14-27

Events: Deadlines: