ML Unit Upgrade: The Machine Learning Unit (Unit 1) has gotten an overhaul. Due to Medium articles being locked behind a paywall, some of the items were outdated. In addition, I wanted to give a more in depth explanation as to how ML algorithms are formulated, with the example being basic linear regression. Please reach out to me if there is anything you see issue with. - Varun Ananth
Notebooks in repo soon obsolete: For ease-of-use and accessibility, we have decided to migrate the
.ipynb files stored in this repo over to Google Colaboratory. This removes the need to set up
an environment and install Anaconda on your personal machine. Colab has many libraries pre-installed and they can be easily imported without much hassle. As of now, only the Unit 1 notebook has been migrated due to some pending updates. The links should all point to Colab soon though. Cheers!
How to use: Take a look at the “Schedule” tab to see how UW I2 uses this website/megadoc. You can take our schedule but since we are on a quarter system it has just 10 weeks of content. All you have to do to run a course like this is create a schedule/syllabus like we have, and have someone able to give a lecture each week on the upcoming unit. All learning + projects are contained on this website!
How to contribute: If you wish to add to this repository of knowledge. please consult the wiki tab of this repo: Website Repository Wiki
Hello and welcome to the Interactive Intelligence Intro to Neuro/AI Course website! This website is designed as a template that shows you how to run a course such as this, as well as resources to do so.
Here is a short story of how this came to be: In the winter of 2022, a group of individuals saw the need for a Neuro/AI club at the University of Washington Seattle (the university did not even have an active ML club at the time). The club, now known as Interactive Intelligence, was formed on the premise of studying and researching on intelligence in systems. Often times this lended itself to philosophical discussions about AI and what AGI would mean for the world. As the club progressed, project groups were formed and within one year the club had presented posters at two conferences and had one paper concerning Reinforcement Learning accepted and published in a journal. Things were picking up steam. As more and more people heard about the club, we had new recruits join. Unfortunately, we had no rigorous methods of education and many people were left to their own devices to learn these difficult Neuro/AI concepts. This led to a lull in motivation from newer members who found the prospect of digesting all this information simply impossible.
This is where the Intro course came in. UW I2 members banded together and in just 2 weeks created the framework for the course you see today. It was refined through a pilot course and by the Spring of 2023, we were able to publish our megadoc on this website. The megadoc is over 40 pages of Neuro/AI content on these topics:
- Linear Regression
- K-Means Clustering
- Neural network “anatomy”
- Backpropagation calculus
- Action potentials
- Major brain modules
- The Visual System
- Primary visual cortex
- Computer Vision
- Convolutional neural networks
- Reinforcement Learning
- Key RL vocabulary
- Q, V, Bellman Equations
- Deep Q Learning
- Epsilon greedy explore-exploit
- The cerebellum
- Language Modeling
- Word Embeddings
- Recurrent neural networks
- Backpropagation through time
- Truncated backpropagation through time
- AI Fairness/Theory
- Pain in RL/Neuroscience/Cognitive Science
- Biological pain in the context of RL
- MaxPain Algorithm
- Networks and Systems of the Brain
- Nature vs. Nurture
- Brain Networks
- Jupyter notebooks
- Self-learning skills
This course is a very basic survey of Neuro/AI and was made to get people excited about intelligence research. We wanted to share this amazing resource with others that wish to use it in their own club. If you need any assistance getting this up and running, please contact the education lead in the “Creators” tab.