Some Idea from Ng Video
1. Gradient descent formula for linear regression and logistic regression are identifcal
except h_theata definition is different. Why the gradient formula are the same?
2. Why go neural net? XOR –> perception or logistic regression can not classify. But neural net can.
So the key is to do island classification?
=> machine learning and topology relationship????
Very interesting topic: neural network, topology and manifold.