Do you know any good software implementations of “classical conditioning”?
That is to say, you can teach it so that a pair of stimuli a short time after each other results in a given response? So the AI learns to associate the two stimuli?
As an example, if I was to program a software simulation of Pavlov’s dog. Is there any good libraries out there that have solved this?
A neural net can do the job. But they aren’t especially fast to train.
Yeah, a neural network will work for this as long as you can collect the data and then train it offline. If you want it to work in realtime in a game without an offline training step, then you would need to fake it using other AI methods.
Neural networks are similar but not the same thing. Neural network is built to work as a classifier. Pavlov response is different.
It might be a good idea to check out information related to Creatures series because they did something like that:
https://en.wikipedia.org/wiki/Creatures_(video_game_series)
Most likely scenario is that in a videogame you’ll have a set of predefined simuli and predefined responses, and will simulate pavlov training by making a 2d grid which will track which, say, two stimuli coincede to each other.
And, of course, you will not be able to make a real dog on a home machine. Not enough computational power.