Day 1 Intro Learning a Line and Artificial Neurons
Today's Notebooks:
Approximate Plan
Machine Learning Toward AI Session 1
-
0-30 mins: Technical Prerequisites — Making sure everyone is on the same page and able to open and run stuff, has access to the repository we will be using, Jupyter and IDEs etc.
-
30-70 mins: Neurons and Lines — Some smartboard chalk and talk from Josh
-
Learning A Line (Regression) vs. Artificial Neurons
-
Changing Perspective: Functions and Equations to Parameters and Learning
-
Loss Function and Training Loop
-
Soapbox about integrating Maths and Programming
-
-
70-90 mins: Notebooks and Tutorials (as much as we get to)
Lunch (30-60 mins depending on how we are feeling)
-
Machine Learning Toward AI Session 2
-
0-45 mins: Notebooks and Tutorials (as much as we get to)
-
45-90 mins: From Neurons to networks
-
Loss Function and Training Loop Intuition: Loss Surface
-
Distance and Loss: from Lists to Mean Squared Error
-
-
90-as long as you want to go: Working on "Learning a Line", Prep for Tomorrow!
Resources
Some things that might be helpful:
-
Pycharm IDE: https://www.jetbrains.com/pycharm/
-
Google Collab: https://colab.research.google.com/
-
Way of getting python, jupyter, sqlite3, and a bunch of libraries on any computer without admin permissions: https://www.edupyter.net/en/
-
3 Blue 1 Brown: Playlist on Neural Networks
-
Welch Labs: Playlist on Neural Networks
-
My Github: https://github.com/deweydex/