What’s on your mind today? Let’s talk about three brain-like characteristics of artificial neural networks. First, there’s the fact that it’s a network of connections with extreme fan-out; in the brain one neuron can connect to ten thousand other neurons. Then all of these connections can have strengths and biases. Finally, there are activations: when the input exceeds some threshold, the output can change discreetly, it can jump. The fact that it’s a network of connections is a big...