When I was young I was uncertain whether the Physical Symbol Systems Hypothesis was a good idea: did this just point at trivial Turing universality, or did it claim that we should build AI in a symbolic paradigm? Now it seems that much of the field suffered the same confusion.
-
-
Replying to @Plinz
Personally I am very deeply concerned by the Symbol Grounding problem and it's not going away any time soon. However, recent work on integrating symbolic and sub-symbolic approaches is promising: https://homes.cs.washington.edu/~pedrod/papers/cacm19.pdf … and http://www.semantic-web-journal.net/content/neural-symbolic-integration-and-semantic-web …. Interested in your thoughts.
1 reply 0 retweets 0 likes -
Replying to @AnneCregan
Symbol grounding is just a specific framing of model grounding. A model is fully ‘grounded’ if it is isomorphic to the system it models. Here, we are looking for a good-enough isomorphism to the universe, i.e. an interactive VR that predicts sensory data.
1 reply 0 retweets 0 likes -
Replying to @Plinz
Do you have a ref for ‘model grounding’ please? Yes, usually isomorphic but also appropriately anchored through specific correspondence point or axes connecting model and reality. These could be physical, like the Greenwich meridian or conceptual.
1 reply 0 retweets 0 likes -
Replying to @AnneCregan
The model is grounded in itself: everything fits to the same unified function that is also compatible with the sensory information. This function is the universe.
1 reply 0 retweets 0 likes -
Replying to @Plinz
I meant just a ref for what you mean by ‘model grounding’, in general please
1 reply 0 retweets 0 likes
I am sure someone somewhere must have written about that, but I don’t know!
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.