Thanks, I tried to understand AI in class and web, but no one simplified the concept like you. At least I think people in my level really learned something.
A note to Better Trading and interested friends here. Actually the loss value cannot be zero eventually. Even if it reaches zero, either the problem is too easy, or it is over-fitting. Thank you for the great preparation!
Regression is a task, while statistics is a branch of maths that is crucial to the study of AI. AI is the study of creating agents that behave intelligently, for example, being able to learn, generalise, create on its own. Therefore, the model in the video is as "AI" as deep learning models.
AI is a massive field containing data analysis, deep learning... Plus, a regression could be a component in Neural Network as well. so, you can say this is a simplified Neural Network
When i update the weight in every new trial, does it mean i have deleted memory of logan's learning experience in the past? how does it store the learning experience?
