Here's a trivial example to illustrate the difference between pattern recognition and reasoning and it impacts behavior generation: let's say you encounter, for the very first time, a glass door with ⅃⅃Uꟼ written on it.
-
-
Only multi-step reasoning (which is slow and resource-intensive) enables zero-shot adaptation to novel situations. Meanwhile, pattern recognition makes navigating well-known situations faster and more efficient.
Show this threadThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
@underthec_ ja dicka interesante :PThanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Wouldn't this still be pattern recognition? You have seen before how a glass door works, and you know that it mirrors the letters on the other side. You are accustomed to 3D space and know how push/pull doors work. Knowing these patterns, you know to push the door instead of pull
-
I think he's saying that reasoning is when there are multiple distinct pattern matching steps that need to be combined in novel ways. Maybe that's also pattern matching. But the space of combined pattern matchers is exponentially bigger.
- Show replies
New conversation -
-
-
Isn't this the example from the book on System 1 (pattern reco) vs System 2 (reasoning)?
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Right, every time we want to teach a robot to open doors somebody needs to hurry and label all of them
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
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.