Course Logistics & Policies

Course Information

Course: Introduction to Machine Learning
Course Numbers: MECH534, EECE490, MECH678, EECE690
Semester: Fall 2026
Credits: 3 credit hours
Prerequisites: Linear algebra, calculus, statistics and probability, programming
Lectures: Tuesday and Thursday, 11:00AM-12:15PM
Location: IOEC 224C

Instructor: Joseph Bakarji

• Email: jb50@aub.edu.lb
• Office: Bechtel 418
• Office Hours: Thursday 1:30-3:30PM (or by appointment)

Teaching Assistants

Summary of Learning Objectives

By the end of this course, successful students will be able to:

1. Understand the mathematical foundations underlying machine learning algorithms
2. Implement core ML algorithms from scratch using Python and NumPy
3. Apply appropriate machine learning techniques to solve real-world problems
4. Evaluate model performance and diagnose common machine learning issues
5. Communicate results effectively through visualizations and technical reports

Grading & Assessment

Setting Up Python

You need a working Python environment for the whole course. Anaconda is the simplest route and gives you the scientific stack in one install.

1. Install Miniconda from docs.conda.io. Miniconda is the small version; full Anaconda works too and is heavier.

2. Make an environment for this course, so nothing you install here breaks anything else on your machine:

   conda create -n intro2ml python=3.11
   conda activate intro2ml
   

3. Install what we use:

   conda install numpy scipy matplotlib pandas scikit-learn jupyter
   

4. Check it works. jupyter notebook should open in your browser, and import numpy should run without complaint.

Activate the environment (conda activate intro2ml) every time you work on the course. If a command is not found, that is usually what you forgot.

If you are new to Python or NumPy, work through the CS231n Python and NumPy tutorial before the second week. It is the fastest way in, and it covers exactly the subset we use: arrays, indexing, broadcasting, and plotting.

Google Colab works if your machine will not cooperate, but a local environment is worth the hour it costs.

Late Policy

• Late days: You get a total of 5 (emergency) late days to use on any assignment submission (except project). You can only use 3 days per assignment. Use them wisely.
• Penalty: 10% penalty per day late beyond credit late days (including weekends)
• Extensions: Only available for well-documented emergencies.

Assignment Policies

Submission Requirements

• Submit on Moodle - link provided on the assignment page
• Include a .ipynb versions of notebooks for code, and a .pdf for the writeup zipped for submission.
• All code must be runnable and well-documented

Collaboration Policy

• Assignments: Discussion of concepts encouraged with your classmates. BUT you must solve and write the problems yourself. You can’t submit duplicate code or writeups. If you’ve used LLMs make sure to state how and where. Cite any external resources used.

• Projects: Collaboration allowed with proper attribution. Groups must be approved by instructor. Individual contributions must be clearly identified in final report and presentation.

Communication

Preferred Communication Methods

For general inquiries, you should use the course Slack channel to ask questions to your classmates and teaching assistants (you get participation points if you ask or answer questions on Slack). If you don’t get an answer, go to the TA office hours when available; if not, come to my office hours. If all fails, email me. In code:

def help_route(slack_answered: bool, ta_office_hours_possible: bool, prof_office_hours_possible: bool) -> str:
    if slack_answered:
        return "Resolved on Slack"
    if ta_office_hours_possible:
        return "Go to TA office hours"
    if prof_office_hours_possible:
        return "Go to instructor office hours"
    return "Email the instructor"

Announcements

• Check course website regularly for updates
• Important announcements will be sent via email or Moodle

Course Success Tips

For assignments, start early - they take longer than expected. Don’t procrastinate on reading and practice problems. Practice Python regularly, even for short periods. When you use LLMs, always write down how you used it and what you got from it. Use debugging tools and read error messages carefully. Don’t rely too much on LLMs. Comment your code thoroughly for later review. Back up your work frequently!

Choose projects aligned with your interests, demonstrate your understanding of the course material, and show your creativity. Start with simple approaches before attempting complex methods. Document everything as you go. Test your code frequently with small examples.

To get the most from the class, come prepared with questions from readings and videos assigned in the schedule for class preparation. Participate actively in discussions and exercises. Review material regularly, don’t cram.

University Policies

This course adheres to all American University of Beirut policies regarding:

• Academic integrity and honor code
• Student conduct and classroom behavior
• Accommodation for students with disabilities
• Non-discrimination and harassment policies

For complete policy details, consult the AUB Student Handbook and academic regulations.

These policies are designed to create a fair, supportive, and productive learning environment. Questions about any policy should be directed to the instructor.

Last Updated: August 29, 2025