Interesting! This is what I was getting at. I'm trying to figure out at what level decisions are made and how are they made. From what I can tell so far, it seems like neurons act as nodes that either fire or not based on the information they receive from previous neurons. The information they receive from previous neurons either encourage or discourage activation, each at a different strength. Once a neuron receives enough encouragement to fire from the previous neurons, it fires and sends its signal to the neurons it is connected to, which they take as encouragement or discouragement. In a sense, decision-making is a series of very long logistic regression models. Each previous neuron serves as a predictive factor with it's own polarity (encouragement or discouragement) and coefficient (magnitude of signal). Learning is changing the values of coefficients so that predictive factors have different impacts on the outcome variable. With this in mind, then at least 2 neurons are needed to make a decision. The more neurons, the more variables can be included in the decision making process. Does that make sense?
you are viewing a single comment's thread
view the rest of the comments
view the rest of the comments
replies: