3 ms·

The Wikipedia article on the Hutter Prize should have a cross-reference to the Loebner Prize as there's a close analogy and some of the same criticism applies,

by bloak 6y ago

The Wikipedia article on the Hutter Prize should have a cross-reference to the Loebner Prize as there's a close analogy and some of the same criticism applies, I think.

maratc 6y ago
> Welcome to Wikipedia, the free encyclopedia that anyone can edit. Hint :)
lifthrasiir 6y ago
That analogy doesn't work. Most criticisms to the Loebner Prize boil down to the human judgement (and agents that exploit this aspect without actual intelligence) and the Hutter Prize doesn't have any. The AI connection in the Huffer Prize is, while still up to debate, at best suggestive and not a goal itself.
- bloak 6y ago
  There are some important differences, and the subjectivity of human judgement is a major one. Also, compression is an interesting goal in its own right. But if the Hutter Prize is proposed as a way of encouraging AI research then I still claim that some of the criticism of the Loebner Prize is applicable. To me it seems doubtful whether compression of a 1 GB text corpus could benefit from AI even in theory: if you can get it down to about 15 MB without AI then any AI would have a very tight budget. Perhaps compression of a 1 TB text corpus could benefit from AI in theory, but that doesn't mean that people working on AI would have much chance of winning the prize in the foreseeable future because there might be lots of non-AI techniques that provide a bigger improvement in the short or medium term. (A similar criticism was made against the Loebner Prize in the 1990s. Sorry I can't immediately find a link to it.)
  - phreeza 6y ago
    The argument is that AI and compression are isomorphic, so in that view there is no such thing as non-AI compression techniques. And in fact if you look at how common compressors work, they build a good model of the next token, given the previous tokens, exactly the same as language models in AI.

by bloak 6y ago

The Wikipedia article on the Hutter Prize should have a cross-reference to the Loebner Prize as there's a close analogy and some of the same criticism applies, I think.

maratc 6y ago

> Welcome to Wikipedia, the free encyclopedia that anyone can edit. Hint :)

lifthrasiir 6y ago

That analogy doesn't work. Most criticisms to the Loebner Prize boil down to the human judgement (and agents that exploit this aspect without actual intelligence) and the Hutter Prize doesn't have any. The AI connection in the Huffer Prize is, while still up to debate, at best suggestive and not a goal itself.

bloak 6y ago
There are some important differences, and the subjectivity of human judgement is a major one. Also, compression is an interesting goal in its own right. But if the Hutter Prize is proposed as a way of encouraging AI research then I still claim that some of the criticism of the Loebner Prize is applicable. To me it seems doubtful whether compression of a 1 GB text corpus could benefit from AI even in theory: if you can get it down to about 15 MB without AI then any AI would have a very tight budget. Perhaps compression of a 1 TB text corpus could benefit from AI in theory, but that doesn't mean that people working on AI would have much chance of winning the prize in the foreseeable future because there might be lots of non-AI techniques that provide a bigger improvement in the short or medium term. (A similar criticism was made against the Loebner Prize in the 1990s. Sorry I can't immediately find a link to it.)
- phreeza 6y ago
  The argument is that AI and compression are isomorphic, so in that view there is no such thing as non-AI compression techniques. And in fact if you look at how common compressors work, they build a good model of the next token, given the previous tokens, exactly the same as language models in AI.