Conceptually, logistic regression is a simple neural network. A neural network with no hidden layers and a softmax output layer using cross-entropy loss is prec

Conceptually, logistic regression is a simple neural network. A neural network with no hidden layers and a softmax output layer using cross-entropy loss is precisely equivalent to logistic regression. Though, if you were to extend your suggestion of logistic regression to include its locally-weighted variants, I'd gladly support your cause.