Connectionists: Early history of symbolic and neural network approaches to AI

[Volker Tresp]

Hi all,

there is no doubt in my mind that the brain has symbols. A symbol might be represented as a single neuron, an ensemble of neurons, or an activation pattern of a set of neurons.

The question is: does the brain use symbolic reasoning (FOL-type)?
Obvious, some of us can do it sometimes!

More in the Tensor Brain paper: https://arxiv.org/abs/2109.13392

Best,
Volker


From: Connectionists <connectionists-bounces at mailman.srv.cs.cmu.edu> on behalf of Gary Marcus <gary.marcus at nyu.edu>
Date: Tuesday, 20. February 2024 at 08:56
To: "Weng, Juyang" <weng at msu.edu>
Cc: "connectionists at mailman.srv.cs.cmu.edu" <connectionists at mailman.srv.cs.cmu.edu>
Subject: Re: Connectionists: Early history of symbolic and neural network approaches to AI

Strawperson alert:

there are not any neurons that have a one-to-one correspondence to the symbol

How about collections of neurons? What do you think logicians do when they manipulate symbols? Is it magic? Parapsychological?



On Feb 19, 2024, at 10:41 PM, Weng, Juyang <weng at msu.edu> wrote:
there are not any neurons that have a one-to-one correspondence to the symbol
-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://mailman.srv.cs.cmu.edu/pipermail/connectionists/attachments/20240220/68d2ea88/attachment.html>