Show this thread
Human perception, even in its most basic forms, involves remarkable amounts of abstraction and reasoning. It is far more than mere pattern recognition.
-
-
Thanks. Twitter will use this to make your timeline better. UndoUndo
-
-
-
This Tweet is unavailable.
-
It’s a sign of our time that we haven’t used realizations like this to make much stronger advances in CS/
#datascience to solve some of the world’s largest & most complex & intractable resource problems. https://cleanapp.io/cleanappculture/ …
End of conversation
-
-
-
This Tweet is unavailable.
-
Agree 100%. This is why we advocate that ML/AI models be trained on huge global data sets concerning waste & “trash” — where humans have as close to universal normative perceptions, with a lot of valuable pluralism that can be
#machinelearned.https://www.change.org/p/want-google-to-activate-trash-hazard-reporting?utm_source=embedded_petition_view …
End of conversation
-
-
-
How do you differentiate abstraction/reasoning from pattern recognition?
-
Abstraction attempts to generate useful new hypotheses about relationships in data; pattern recognition tries to group/classify relationships in data. Deep learning can do certain types of pattern recognition given sufficient data, but can't [yet?] do abstraction/generalization.
- Show replies
New conversation -
-
-
Time for Deep Learning folks to dig up old AI and Cybernetics literature. It's pure hubris to think they've discovered the path to general intelligence without understanding previous research.
End of conversation
-
-
-
abstraction and reasoning = higher order patterns ?
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.