no code implementations • 24 Jul 2020 • Marissa Masden, Dev Sinha
Informed by the basic geometry underlying feed forward neural networks, we initialize the weights of the first layer of a neural network using the linear discriminants which best distinguish individual classes.
