Online Session Locator

Session Description

Ticketed event: $30.00 advanced registration and $40.00 on site registration
In this interactive 2.5-hour hands-on workshop, participants will learn how to build a complete Tic Tac Toe game in MATLAB, culminating in an AI-powered computer opponent that improves its play over time using reinforcement learning (Q-learning). The first half of the workshop will walk through graphical design, game logic, and implementing a learning agent from scratch. The second half, led by experts from MathWorks, will showcase additional AI/ML tools and workflows available in MATLAB to accelerate student learning and project-based instruction.

Attendees will leave with complete code, instructional materials, and ideas for using games to teach AI, logic, and programming fundamentals in the classroom.
------
Materials to Provide Attendees:
-Starter Live Script (pre-built grid, stubs for code)
-Pre-trained QTable.mat file (for fast demos)
-Full working solution with RL enabled
-Slide deck (5–7 slides) with Q-learning overview
-Link to MATLAB Online if no local install
-Optional: printed or PDF handout summarizing the algorithm flow
------
Optional Add-Ons (if time or interest allows):
-Switching between random and trained AI
-Adding a GUI with buttons (instead of ginput)
-Letting the AI play against itself (self-play trainer)
-Showing metrics like win rate over time
------
Agenda & Outline:
Time - Topic - Details
--
0:00–0:10 Welcome & Setup
-Brief introduction, learning goals
-Ensure everyone has MATLAB or MATLAB Online access
--
0:10–0:30 Building the Game Board in MATLAB
-Creating a 3×3 grid
-User input via ginput
-Drawing Xs and Os
-Storing board state in matrices
--
0:30–0:45 Adding Game Logic: Turns, Win Checking
-Alternating turns
-Validating moves
-Detecting wins/draws
-Using matrix sums to check for victory
--
0:45–1:05 Introducing the Computer Opponent (Random)
-Letting the computer make random moves
-Generating legal random positions
-Enhancing gameplay loop
--
1:05–1:25 From Random to Smart: Implementing Q-Learning
-Introduction to Q-learning
-Representing state-action values
-Epsilon-greedy strategy
-Updating Q-values
--
1:25–1:30 Save & Load AI Memory (Q-table)
-Storing Q-tables in .mat files
-Persistent learning between games
--
1:30–1:45 Intersession Break (15 min) Bathroom/stretch/networking
--
1:45–2:15 Expanding to the Classroom: AI Tools in MATLAB (Led by MathWorks personnel)
-Reinforcement Learning Toolbox
-Training visualizations
-MATLAB Online & Grader
-Deployable AI demos
--
2:15–2:30 Wrap-Up & Q&A
-Review goals & outcomes
-Showcase a well-trained AI
-Q&A
-Share links to materials (Live Scripts, code, videos)

Speakers

1. Dr. Stephen Andrew Wilkerson P.E.

   York College of Pennsylvania

   Stephen Wilkerson received his Ph.D. in Mechanical Engineering from Johns Hopkins University in 1990. He spent 33 years at the Army Research Laboratory (ARL) at Aberdeen Proving Ground, with the last 15 years of his career devoted to advancing robotic and drone systems. Before retiring in 2016, he served as an exchange scientist in Germany in 1993 and was a full-time faculty member at the United States Military Academy (USMA) at West Point from 1996 to 1997, and again from 2010 to 2012. He has also held adjunct faculty positions at Harford Community College, Towson University, and the University of Maryland, Baltimore County (UMBC).

   Dr. Wilkerson is currently an Associate Professor at York College of Pennsylvania, where his research centers on drone technologies for agricultural applications. His work emphasizes the development of AI-driven solutions to enhance drone operations and improve multispectral analysis of farm fields.

   Beyond his research, he remains deeply committed to STEM education and outreach, with a strong focus on inspiring and preparing the next generation of engineers and scientists.

2. Dr. Elvira Osuna-Highley

   MathWorks

   Dr. Elvira Osuna-Highley is part of a global team supporting academic research and teaching at MathWorks. Before joining MathWorks, she was on the faculty of the Computational Biology Department at Carnegie Mellon University. She holds a doctorate in Biomedical Engineering from Carnegie Mellon University, where her research involved applying machine learning techniques to fluorescence microscope images.

3. Gen Sasaki

   MathWorks

   Gen Sasaki is a Principal Customer Success Engineer at the MathWorks, working to ensure university educators and students get the most out of MATLAB. He holds a BSME and MSME with a focus on control systems. He in automotive and aerospace applications for nearly 30 years, in powertrain, various embedded controls, and functional safety.

4. Dr. Scott F. Kiefer

   North Carolina State University at Raleigh

   Scott Kiefer is an Associate Teaching Professor at Mechanical and Aerospace Engineering at North Carolina State University. He can be reached at sfkiefe2@ncsu.edu