Arora suggests citing John Rupert Firth’s distributional semantics instead
Unfortunately doesn’t make sense to cite PI here. The argument Wittgenstein makes in PI is superficially related to the idea that the same set of symbols can have multiple meanings, but one could certainly reject every argument W makes and still believe that two different words
oh so you're a ml researcher who likes philosophy? well have you ever cited one in an ML paper?
Negative users criticize the ML researcher's advice against citing Wittgenstein's PI for discouraging deeper philosophical engagement by machine learning researchers.
No Digg Deeper questions have been answered for this story yet.
can happen to have the same symbols that represent it. After all, the later is an easily observable fact about (most?) human languages.
In classical Latin, malum could mean “apple” or “evil.” In Spanish, vino can mean “wine” or “she came.”
Unfortunately doesn’t make sense to cite PI here. The argument Wittgenstein makes in PI is superficially related to the idea that the same set of symbols can have multiple meanings, but one could certainly reject every argument W makes and still believe that two different words
Engagement with philosophy (and other disciplines in general) is a good thing that machine learning researchers should strive to do more of. But superficial engagement like this just underlines how little ML researchers pay attention to work in other disciplines
can happen to have the same symbols that represent it. After all, the later is an easily observable fact about (most?) human languages.
In classical Latin, malum could mean “apple” or “evil.” In Spanish, vino can mean “wine” or “she came.”
And that’s without even opening the can of worms of whether “computer virus” is even two words or if it’s one word with a space character in it.
Engagement with philosophy (and other disciplines in general) is a good thing that machine learning researchers should strive to do more of. But superficial engagement like this just underlines how little ML researchers pay attention to work in other disciplines
“Virus” is obviously an example of a polysemous word and not an example of a homonym, the way I portrayed originally in this thread. I’m not sure what went wrong in my brain that caused me to complete these things, but it also doesn’t bear on if the citation is appropriate
Unfortunately doesn’t make sense to cite PI here. The argument Wittgenstein makes in PI is superficially related to the idea that the same set of symbols can have multiple meanings, but one could certainly reject every argument W makes and still believe that two different words
@BlancheMinerva yeah should have cited Firth on distributional semantics if anything
Unfortunately doesn’t make sense to cite PI here. The argument Wittgenstein makes in PI is superficially related to the idea that the same set of symbols can have multiple meanings, but one could certainly reject every argument W makes and still believe that two different words