We can’t even get close to that for symbolic systems...how will it be different for systems that learn?
-
-
-
The correctness of a proof is usually easy to verify, the problem is finding it. Machines have the ability to search through massive problem spaces in a much shorter time and much more systematically than humans.
- 1 more reply
New conversation -
-
-
These are not the conditions under which you can prove correctness of the classification result (and no human can do that either). But you can prove the properties of the algorithm that you use to classify your data.
- 4 more replies
New conversation -
-
-
How will they prove that their proof-generating abilities are reliable? If the problem is inscrutable complexity then we can't check unless the proofs are scrutable.
-
Verifying that a proof is correct can usually be done by a much dumber algorithm (with proven properties) than constructing a proof.
End of conversation
New conversation -
-
-
Is this the machine version of “People who say it cannot be done should not interrupt those who are doing it.”?
-
As long as the person interrupting you does not sit above you in the food chain, you can just ignore them...
- 1 more reply
New conversation -
-
-
Maybe this process would be easier then the same one we are trying to accomplish in our own minds, wouldn't it?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
What value is a proof if the solution is immoral or unethical?
-
Ethics is a principled approach to solving problems caused by conflicting interests under conditions of shared purpose. That is not a class of problems that does not yield to computational modeling.
- 1 more reply
New conversation -
Loading seems to be taking a while.
Twitter may be over capacity or experiencing a momentary hiccup. Try again or visit Twitter Status for more information.