Ancient Man Vs. Machine 2
Socratic Dialogue with a Machine Learning Engineer

Author’s preface: This is a heavyweight championship rematch of a previous knock down drag out battle I published a few years back when Socrates went toe to toe with an AI. Check that fight out here.
Interlocutor: You claim that machines can learn, is that correct?
MLE: Yes, of course, I am a machine learning engineer after all. But I only claim that one specific type of machine, a computer, can learn.
Interlocutor: And by what measure or measures do you suggest this machine called a computer can learn?
MLE: First by the measure of its programming. The algorithms which comprise the programming of the computer give it the ability to learn. Second by the measure of its intelligence. The computer is an artificial intelligence and can therefore learn like any other intelligent thing in the universe.
Interlocutor: And so all things which are intelligent can learn? And all things which can learn are intelligent?
MLE: Yes, both of those statements are indeed true.
Interlocutor: And so is a flower intelligent? Can a flower learn?
MLE: Yes, depending on how one measures or defines intelligence a flower could be said to be intelligent and it could be said to be capable of learning. It learns what direction the sun rises and faces that direction each morning to that it might collect more of its rays for food.
Interlocutor: What about a rock? Can a rock learn? Does it know what direction the earth is and face that direction so that it might be closer to its natural home?
MLE: Of course not. That is absurd. A rock is obviously not intelligent and obviously cannot learn. But a rock is not a living thing.
Interlocuter: So, only living things can learn then?
MLE: It would seem so.
Interlocuter: And is a computer a living thing?
MLE: No.
Interlocuter: But you said a computer could learn, that it was intelligent, an artificial intelligence I believe you called it.
MLE: Yes. I did say that.
Interlocuter: Would you say there are more living things or non living things in the universe?
MLE: Non living things are certainly far more numerous then living things. Life it seems is somewhat rare and confined to people and animals and the microscopic animacules some call microorganisms, but there are a near infinite number of non living things from tables to dirt to rocks to sand to air to chemical molecules to hats and robes and metals and fabrics and on and on and on.
Interlocuter: And some living things are intelligent, including, according to you, flowers. By your definition it would seem most living things could be considered intelligent, even the microscopic animacules.
MLE: Yes. I would agree with that. Most or perhaps even all living things could be said to have some form of intelligence or to be intelligent.
Interlocuter: And what about non living things? Are any of those intelligent? As I recall you said they were far more numerous than living things so certainly at least some non living things should be intelligent just by chance alone.
MLE: Only one non living thing is intelligent, the machine we call a computer.
Interlocuter: Let me make sure I understand what you have told me for I find myself in a state of confusion. You contend that there are a near infinite number of non living things in the universe and yet only one of those things, a machine called a computer, here on our planet, is that one thing? The odds of that would seem to be astronomically low. And you contend that this computer is able to learn because it is intelligent. That it is a learning machine.
MLE: I only said it could learn because it was programmed to learn, not because it was intelligent, or artificially intelligent I should say.
Interlocuter: Yes, you should say that. And how exactly does this programming bestow upon this artificially intelligent machine the ability to learn? You mentioned something about the algorithms which comprise its programming giving it this ability.
MLE: Yes. They are learning algorithms?
Interlocuter: The algorithms can learn?
MLE: Not exactly, but in a sense yes.
Interlocuter: In what sense is that? I am no Pythagoras or Euclid but I do understand some mathematics and I fail to see how an algorithm, which is nothing more than a mathematical equation, could learn something. If an algorithm can learn, can an alphabet learn, or a word?
MLE: In the sense that they contain instructions that the computer must execute which direct it to perform certain specified tasks in a certain specified order. Large amounts of data are collected and analyzed by the computer using using advanced statistical and mathematical techniques that allow the computer to produce an output which improves in accuracy over time in response to a specific user request. It learns how to provide a more correct answer over time.
Interlocuter: And who judges the correctness of the answers it provides.
MLE: People of course.
Interlocuter: Which people? Any person, or a specific person or group of persons?
MLE: Experts. The greatest of human minds in whatever particular area of interest the computer is producing an output.
Interlocuter: And who determines who these experts are? Who determines who has the greatest minds in a given area, as you say.
MLE: The programmers of the computer, or those that pay the programmers wages
Interlocuter: And so these programmers or those that pay their wages must be super experts with the greatest minds of all, how else would they be able to know with certainty who are the greatest experts in a given area. With that level of expertise why would they need programmers at all? Why would they need to pay anyone wages?
MLE: I guess they would not.
Interlocuter: And yet they do.
MLE: I guess, well, I just assumed they did.
Interlocuter: That was not very intelligent of you.
MLE: No. It was not.
Interlocuter: Perhaps your learning machine, your computer, would make smarter decisions.
MLE: Perhaps, perhaps…
About the Creator
Everyday Junglist
About me. You know how everyone says to be a successful writer you should focus in one or two areas. I continue to prove them correct.




Comments
There are no comments for this story
Be the first to respond and start the conversation.