Explaining Feedforward, Backpropagation and Optimization: The Math Explained Clearly with Visualizations. I took the time to write this long article (>5k words), and I hope it helps someone understand neural networks better.
Link: <a title="Neural Networks: Feedforward and Backpropagation Explained & Optimization" href=" removed link " target="true">mlfromscratch / neural networks explained
I have been studying Machine Learning in the last few months, and I wanted to really get to understand everything that goes on in a basic neural network (excluding the many architectures). Therefore, I took the time to write this long article, to explain what I have learned. In particular, the post on purpose very extensive and goes into the smaller details; this is to have everything in one place. As the site says, it is machine learning from scratch, and I share what I have learned.
The particular reason for posting here, is that I hope someone else could learn from this. The goal is to share the knowledge in the easiest absorbable way possible. I tried to visualize much of the process going on in neural networks, but I also went through the math, to the detail of the partial derivatives.
This was quite a journey, and it took about 1 month to read all the things I have read, and write it down, have it make sense and creating the graphics.
Any constructive feedback is appreciated.