Matrix and Vector Operations in Logistic Regression

Vectorized Logistic Regression

The underlying math behind any Artificial Neural Network (ANN) algorithm can be overwhelming to understand. Moreover, the matrix and vector operations used to represent feed-forward and back-propagation computations during batch training of the model can add to the comprehension overload. While succinct matrix and vector notations make sense, peeling through such notations down to subtle working details of such matrix operations would bring more clarity. I realized that the best way to …

algorithm ann artificial logistic regression machine learning math matrix network neural network operations overload propagation regression the matrix training vector vectorization