<aside> 🛠

Under construction: Check back after Oct 31 for the updated version!

</aside>

This self-paced course is designed for those who:

• Don’t have any technical background beyond high school math
• Need enough understanding about how AI works to evaluate the robustness of safety techniques

Key Takeaways

• A neural network has its reward function optimised through gradient descent
• The main drivers of AI development: algorithms, data, compute
• Understand how LLMs are trained and fine-tuned

Units

1: How does AI learn

Resources

• https://bluedot.org/blog/how-does-ai-learn
• https://www.youtube.com/watch?v=aircAruvnKk&utm_source=bluedot-impact
• https://www.youtube.com/watch?v=IHZwWFHWa-w&utm_source=bluedot-impact
• https://www.youtube.com/watch?v=zjkBMFhNj_g&utm_source=bluedot-impact
• https://medium.com/@richardcngo/visualizing-the-deep-learning-revolution-722098eb9c5?utm_source=bluedot-impact

Bonus maths exercises

These exercises allow you to practice the maths behind reinforcement learning, gradient descent and backpropagation.

• Exercises