= 27d567bc2d48d9f40e9ccc20ea370b15)
jsteinhardt comments on Great Explanations - Less Wrong
You are viewing a comment permalink. View the original post to see all comments and the full post content.
= 783df68a0f980790206b9ea87794c5b6)
Loading…= b715d02e85f7b3b4ae65ca3d3e450868)
Subscribe to RSS Feed
Powered by Reddit
= f037147d6e6c911a85753b9abdedda8d)
You are viewing a comment permalink. View the original post to see all comments and the full post content.
Comments (113)
I think it might be helpful for you to taboo the word "real understanding". It seems like a lot of the disagreement stems from luke and wedrifid being unable to understand what you mean when you use that phrase. To be honest, while I agree with many of your points, I also don't think I understand what "real understanding" is supposed to mean. Perhaps you could restate your original point without use of the word "understanding"?
Even understanding isn't well-defined. I at least don't know of any agreed upon conceptual or mathematical definition. Does Wolfram Alpha understand math? Does a lookup table understand anything?
I think that you really understand a subject if you are able to transfer, teach and artificially recreate the skill or heuristic that is necessary to make useful predictions about the subject. The skill or heuristic needs to be utilizable given limited resources. You also have to be able to abstract and generalize your knowledge about the subject to solve new problems that are not similar to problems previously encountered. Furthermore, you have to be able to prove fundamental mathematical statements and relationships about the subject.
An example would be the ability to teach the game of Go to other agents, transfer your knowledge by writing books on how to play it, and create a Go AI that can play the game by predicting the success of its strategies.