Intuitive guide to handling non-linearity with neural networks.
Excellent. Lucidly put. Keep it up.
That's only the case when the task is binary classification, right?
No no, the pattern holds for other classification tasks as well. I used a binary example for simplicity. If you have three classes, instead of having a single line-separability, you may have a decision boundary like this one: https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2F551c4253-1046-4b20-b8c8-c823f746017b_2736x2840.png
Excellent. Lucidly put. Keep it up.
That's only the case when the task is binary classification, right?
No no, the pattern holds for other classification tasks as well. I used a binary example for simplicity. If you have three classes, instead of having a single line-separability, you may have a decision boundary like this one: https://substackcdn.com/image/fetch/f_auto,q_auto:good,fl_progressive:steep/https%3A%2F%2Fbucketeer-e05bbc84-baa3-437e-9518-adb32be77984.s3.amazonaws.com%2Fpublic%2Fimages%2F551c4253-1046-4b20-b8c8-c823f746017b_2736x2840.png