In my own quest to keep the pace with AI innovations, I'm making this repo in which I illustrate one concept at a time, from undergrad algebra to GPT, image generation and reinforcement learning. Every notebook contains a description of the introduced concept, and a list of previous notebooks required to understand it.
Some of the concepts like derivatives might look too obvious, but if starting from scratch, trusting the process and starting from the most basic concepts might be a good idea.
Pending topics:
- Optimization Algorithms
- Activation functions
- Autoencoders
- Reinforcement Learning
- Hyperparameter Tuning
- Transfer Learning
- Generative Adversarial Networks (GANs)
- Model Deployment
- Alignment
- Graph Neural Networks