Resource type
Use type
Tools
Project name
Standards & Guidelines
Lesson Overview
This lesson introduces students to how GANs work as a result of the interplay between generator and discriminator neural networks. Students will learn how the generator and discriminator compete with one another to generate text, images, videos, and more.
Total Lesson Time: 2-3 days, 45 min each
Learning Objectives: Students will be able to . . .
- Given a word bank, students will explain how a GAN and its components - a discriminator and a generator network - work.
- Describe the relationship between a generator and discriminator in a GAN in their own words
Vocabulary Introduced: generator, discriminator
Pacing:
Day 1
- Opening (5 min)
- Introduction to new material and terms (5 min)
- Generator vs. Discriminator game, whole class (25 min)
- Wrap-up + Exit Ticket (if not continuing to Day 2) (15 min)
Days 2 - 3 Optional, Activities can space multiple days
- Review (5 min)
- Independent work, GANs Exploration! (5 min)
- Generator vs. Discriminator game, student pairs (25 min)
- Wrap-up + Exit Ticket (15 min)
Planning Guide
Preparation Needed: 15-20 min
Prep Needed for Teaching In-Person:
- Print the board game version of the Generator vs. Discriminator Game
- Print the Discriminator Input Data Set
Log in or register to view attachments and related links, and/or join the discussion. If you are already logged in, scroll to the bottom of this page for the links.
Resources
Student Materials:
- GANs Exploration Packet*
- Generator vs Discriminator Printable Game
*Note: GANs Exploration is a long packet: 5 GANs to explore, 2 summative activities to test what they have learned. It could take students two class periods to complete the entire slide deck / packet.
Activity Usage
Copyright held by MIT STEP Lab
License: CC-BY-NC under Creative Commons
These materials are licensed as CC-BY-NC 4.0 International under creative commons. (For more information visit https://creativecommons.org/licenses/by-nc/4.0/). This license allows you to remix, tweak, and build upon these materials non-commercially as long as you include acknowledgement to the creators. Derivative works should include acknowledgment but do not have to be licensed as CC-BY-NC. People interested in using this work for for-profit commercial purposes should reach out to Irene Lee at [email protected] for information as to how to proceed. Attribution-NonCommercial 4.0 International (CC BY-NC 4.0)
Attribution:
The original "How GANs work" activity was created as part of the "Creativity and AI" curriculum by Daniella DePaola and Safinah Ali with support from the MIT Media Lab Personal Robots Group. See more: Exploring Generative Models with Middle School Students
The online google slide deck version of the GAN game was developed by Kate Moore for MIT STEP Lab.
Log in or register to view attachments and related links, and/or join the discussion