4 ms·

While I mostly agree with Michael Jordan, I think that he overstates his argument a small bit. I feel comfortable applying the "AI" label to projects if they: 1

While I mostly agree with Michael Jordan, I think that he overstates his argument a small bit. I feel comfortable applying the "AI" label to projects if they: 1) achieve human level of performance or better. 2) They use any or all of standard techniques like deep learning, NLP, knowledge representation, reinforcement learning, etc. I am just about to hit my 40 year anniversary for getting paid to be doing "AI" related work, with a lenient definition of AI. On the other hand, Artificial General Intelligence is a much stronger term, and is something that I don't expect to see in my lifetime.

bobthechef 6y ago
> 2) They use any or all of standard techniques like deep learning, NLP, knowledge representation, reinforcement learning, etc. But why that? There isn't a principled reason for singling those out. It's just a matter of convention that we call those AI. Pretend you never heard of AI before and you came across these algorithms. No amount of analysis or poking around would lead you to think "oh, man, this is artificial intelligence!". No, you'd see a bunch of statistical techniques that do X, Y, and Z, none of which would jump out at you as "wow, this is AI". All that computers give us is the ability to run tedious computations over larger data sets that would be impractical for human beings to perform. "AI" is what we read into these things. There really is no such thing as such.
- 256lie 6y ago
  What would a principled reason for association look like beyond mere convention? Language is used by different groups to mean different things. Machine learning, logic, control, robotics, linguistics, and cognitive science were publishing in artificial intelligence venues decades ago. Now AI seems to just mean DL/RL.
hervature 6y ago
Serious question, are calculators AI because they perform arithmetic much better than humans and use standard techniques like Taylor series expansion and numerical methods? I feel like you want to add that the technique has to use data as a core part of its algorithm that represents “learning”. But then we’re just back to machine learning that beats humans at a specific task. Is there a need for this term? We already have “superhuman algorithm”. AI has more syllables and we could also use SA which could double as State of the Art.

bobthechef 6y ago

> 2) They use any or all of standard techniques like deep learning, NLP, knowledge representation, reinforcement learning, etc. But why that? There isn't a principled reason for singling those out. It's just a matter of convention that we call those AI. Pretend you never heard of AI before and you came across these algorithms. No amount of analysis or poking around would lead you to think "oh, man, this is artificial intelligence!". No, you'd see a bunch of statistical techniques that do X, Y, and Z, none of which would jump out at you as "wow, this is AI". All that computers give us is the ability to run tedious computations over larger data sets that would be impractical for human beings to perform. "AI" is what we read into these things. There really is no such thing as such.

256lie 6y ago
What would a principled reason for association look like beyond mere convention? Language is used by different groups to mean different things. Machine learning, logic, control, robotics, linguistics, and cognitive science were publishing in artificial intelligence venues decades ago. Now AI seems to just mean DL/RL.

hervature 6y ago

Serious question, are calculators AI because they perform arithmetic much better than humans and use standard techniques like Taylor series expansion and numerical methods? I feel like you want to add that the technique has to use data as a core part of its algorithm that represents “learning”. But then we’re just back to machine learning that beats humans at a specific task. Is there a need for this term? We already have “superhuman algorithm”. AI has more syllables and we could also use SA which could double as State of the Art.