We should explore our natural world, but at what point do we exploit the strategies we have learned in exchange for the exploration. Reinforcement is the neurological and psychological concept of strengthening a reaction to a stimulus, analogous to how a structural engineer would use some physical reinforcement to strengthen various parts of a bridge. Positive reinforcement is a positive reward for what 'trainer' deems positive, and negative reinforcement is this negated, negative reward for what the 'trainer' deems negative. Pavlov's dogs is one research experiment that is fundamental in this science. Reinforcement is the response from the environment that happens in our brains, we create, and reinforce patterns, behaviors, responses, and our methodologies into our minds that manifest into a designed outcome. I say designed because that is simply the best word for it, for there must be a sculptor for there to be a statue. That sculptor is partly your parents, your friends, the shows and movies you watch, or your favorite teacher from high school. They might shape the way you act, dress, talk, or even think. I believe humans are wired, sculpted, to attach to a group that makes them feel wanted and heard. We consider this the purpose of living, attachment, the feeling of wholeness, and shape our lives around it. Accordingly, we will strengthen the neurological pathways, and more abstractly our phycological pathways, that lead to the reactions we wish to elicit.
We often do not think about our minds, think about thinking, think about how our minds came to be, we just assume that it is as it always was and there is not much else there. Do we not wish to think of our minds as a sculpture? Admitting our mind is beautiful, complex, confusing but somehow works perfectly, leads to more questions than answers. Is there a creator or is it natural selection? I believe that it is somewhere in-between these two, there is more beauty than is perceived in science, and it is because of our imperfections as humans that we are not able to see.
Modeled after psychological reinforcement, reinforcement learning is a sub science of artificial intelligence in which an agent will learn an optimal solution through exploring its environment, discovering which moves cause it to receive a positive reward, and exploiting those strategies. It might be helpful to image a maze as an environment for an agent, because the concept of exploring can be quite literal. In games like chess however, it is a bit more complex of a methodology, but the concept of exploration and exploitation remains. The point here is that while reinforcement learning can look really complicated and mathematically convoluted, we can still determine the concepts integral to learning in this setting.