Teaching Calendar 2024-2025

Daily Review

  • 4/8/2025 (Tuesday)
    • Auto-Encoders!
    • Side quest: Check out my app here for turning in Java code. Log in with your credentials for class and submit to the Binary Tree assignment. If you need a file to submit, here’s one that should work
    • Assignment: Create an auto-encoder for the Fashion MNIST dataset. If you get that working, add noise or dropout to the input images and see how well it reconstructs.
  • 4/10/2025 (Thursday)
    • Notes on genetic algorithms and symbolic regression
    • Social Hour
  • Spring Break
  • 4/21/2025 (monday)
    • Help me test javadrop.io by going to javadrop.io or the same thing here
      • You should be asked for a new password. Don’t forget it!
      • Navigate to AP FRQ Practice and Select the “Candy” FRQ
      • Test out the interface and submit a solution. Here’s one that seems to work. Feel free to change it or submit your own.
      • I’m testing the viability of this server to handle concurrent API requests to OpenAI, so everybody should submit their responses within the same minute or so (not all at the exact same time)
      • If it works you should get feedback in 10-15 seconds
      • Feel free to explore/test/poke other parts of the site
    • Research topics
      • Make a copy of this research notebook template in your own google drive
      • Discuss current research ideas and begin to fill out research plan
      • Do some background research. Find 10 relevant papers and add them to Zotero
      • (I think you downloaded Zotero a while ago?)
      • Begin work on annotated bibliography
  • 4/23/2025 (wednesday)
  • 4/25/2025 (friday)
    • Relativistic dynamics of space near a black hole
    • Hawking radiation
    • Homework: Can information be destroyed by a gravitational singularity?
  • 4/29 (Tuesday)
    • Variational Autoencoders
    • See notes and code here
    • Implement solutions in colab or jupyter, etc.
    • here’s a fix
  • 5/5 (Monday)
    • Generative Adversarial Networks
    • Code up solution from handout and generate Fashion MNIST data
    • Note: the provided code is not compatible with Keras 3! Here’s a fix
  • 5/7 (Wednesday)
    • Continue GANs
    • Fix code from last class
    • Implement a Deep Convolutional GAN solution to “FaceGAN”
    • Find your own face data. There are HUGE databases but I’m using 10,000 images of size 256x256 on the 4090 and that’s pretty much a reasonable size. So don’t get too much. Here’s some helper code to wrangle your data into folders
  • 5/9 (Friday)
    • AP something. APUSH?
  • 5/11 (Tuesday)
  • 5/13 (Thursday)