<html><head><meta http-equiv="content-type" content="text/html; charset=utf-8"></head><body dir="auto"><div dir="ltr"></div><div dir="ltr">Dear Stevan,</div><div dir="ltr"><br></div><div dir="ltr">You can define things as you like, but too strict a reading on 4 and 5 leads to weird places. Yes, a literal Turing Tape doesn’t care what my bits are, but I can build algorithms in higher-level languages, or even in low-level hardware, that are sensitive to the structure of the symbols that I am passing. For example, programmers often use signed representations for integers, in which the leftmost bit in the representation of an integer is used to represent whether the number is positive or negative and then further computation (eg comparison at the hardware or software level) is condition thereon [eg <a href="https://www.tutorialspoint.com/signed-binary-integers">https://www.tutorialspoint.com/signed-binary-integers</a>]. In this way, the representation of individual binary integers is far from arbitrary. I would not wish to conclude that they are therefore not symbols, nor that, eg., the comparison I would then between them is not computation or not algorithmic. (And of course the representations of the magnitude of the integer is perfectly orderly, the antithesis ofarbitrary)</div><div dir="ltr"><br></div><div dir="ltr">All of this can be mapped onto a Turing tape, but it doesn’t mean that the microprocessor opcodes or programming instructions that manipulate numbers with reference to the sign bit aren’t manipulating symbols or aren’t algorithmic. It means that we need ways of talking about symbols that are composable, potentially in meaningful ways.</div><div dir="ltr"><br></div><div dir="ltr">I take embeddings to be a very useful extension of that idea, of composed symbols, in which you wind up with a set of stable encodings in which the individual bits are not in fact arbitrary. You still wind up with a (useful) string of bits to represent something. (Binary or otherwise). But you can leverage those bits in interesting ways.</div><div dir="ltr"><br></div><div dir="ltr">Yes, you can do whatever you wind up with in a cumbersome fashion on a Turing Tape, but so what? That fact alone doesn’t tell us all that much about the nature of the computations we want to perform, since (resource and timing limits aside) there is nothing on the table or seriously envisioned that couldn’t be mapped onto the infinite space of things that could be so mapped.</div><div dir="ltr"><br></div><div dir="ltr">Gary</div><div dir="ltr"><br></div><div dir="ltr"><br></div><div dir="ltr"><br></div><div dir="ltr"><blockquote type="cite">On Feb 6, 2022, at 05:48, Stevan Harnad <harnad@ecs.soton.ac.uk> wrote:<br><br></blockquote></div><blockquote type="cite"><div dir="ltr">
<meta http-equiv="Content-Type" content="text/html; charset=utf-8">
<meta name="Generator" content="Microsoft Word 15 (filtered medium)">
<!--[if !mso]><style>v\:* {behavior:url(#default#VML);}
o\:* {behavior:url(#default#VML);}
w\:* {behavior:url(#default#VML);}
.shape {behavior:url(#default#VML);}
</style><![endif]--><style><!--
/* Font Definitions */
@font-face
{font-family:"Cambria Math";
panose-1:2 4 5 3 5 4 6 3 2 4;}
@font-face
{font-family:Calibri;
panose-1:2 15 5 2 2 2 4 3 2 4;}
@font-face
{font-family:"Times New Roman \(Body CS\)";
panose-1:2 11 6 4 2 2 2 2 2 4;}
@font-face
{font-family:Consolas;
panose-1:2 11 6 9 2 2 4 3 2 4;}
/* Style Definitions */
p.MsoNormal, li.MsoNormal, div.MsoNormal
{margin:0cm;
font-size:11.0pt;
font-family:"Calibri",sans-serif;}
a:link, span.MsoHyperlink
{mso-style-priority:99;
color:blue;
text-decoration:underline;}
pre
{mso-style-priority:99;
mso-style-link:"HTML Preformatted Char";
margin:0cm;
font-size:10.0pt;
font-family:"Courier New";}
span.HTMLPreformattedChar
{mso-style-name:"HTML Preformatted Char";
mso-style-priority:99;
mso-style-link:"HTML Preformatted";
font-family:Consolas;}
span.EmailStyle23
{mso-style-type:personal-reply;
font-family:"Calibri",sans-serif;
color:windowtext;}
.MsoChpDefault
{mso-style-type:export-only;
font-size:10.0pt;}
@page WordSection1
{size:612.0pt 792.0pt;
margin:72.0pt 72.0pt 72.0pt 72.0pt;}
div.WordSection1
{page:WordSection1;}
--></style><!--[if gte mso 9]><xml>
<o:shapedefaults v:ext="edit" spidmax="1026" />
</xml><![endif]--><!--[if gte mso 9]><xml>
<o:shapelayout v:ext="edit">
<o:idmap v:ext="edit" data="1" />
</o:shapelayout></xml><![endif]-->
<div class="WordSection1">
<p class="MsoNormal">0. It might help if we stop “cognitizing” computation and symbols.
<span style="font-size:12.0pt"><o:p></o:p></span></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">1. Computation is not a subset of AI. <o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">2. AI (whether “symbolic” AI or “connectionist’ AI) is an application of computation to cogsci.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">3. Computation is the manipulation of symbols based on formal rules (algorithms).<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">4. Symbols are objects or states whose physical “shape” is arbitrary in relation to what they can be used and interpreted as referring to.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">5. An algorithm (executable physically as a Turing Machine) manipulates symbols based on their (arbitrary) shapes,
<i>not their interpretations</i> (if any).<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">6. The algorithms of interest in computation are those that have at least one meaningful interpretation.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">7. Examples of symbol shapes are numbers (<i>1, 2, 3</i>), words (<i>one, two, three</i>; onyx, tool, threnody), or any object or state that is used as a symbol by a Turing Machine that is executing an algorithm (symbol-manipulation rules).<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">8. Neither a sensorimotor feature of an object in the world, nor a sensorimotor feature-detector of a robot interacting with the world, is a symbol (except in the trivial sense that any arbitrary shape can be used as a symbol).<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">9. What sensorimotor features and sensorimotor feature-detectors (whether “symbolic” or “connectionist”) might be good for is connecting symbols inside symbol systems (e.g., robots) to the objects that they can be interpreted as referring
to.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">10. If you are interpreting “symbol” in a wider sense than this formal, literal one, then you are closer to lit-crit than cogsci.<o:p></o:p></p>
<p class="MsoNormal"><o:p> </o:p></p>
<p class="MsoNormal">Stevan Harnad<o:p></o:p></p>
<p class="MsoNormal"><span style="font-size:14.0pt"><o:p> </o:p></span></p>
<p class="MsoNormal"><span lang="EN-US" style="font-size:14.0pt"><o:p> </o:p></span></p>
<div style="border:none;border-top:solid #B5C4DF 1.0pt;padding:3.0pt 0cm 0cm 0cm">
<p class="MsoNormal" style="mso-margin-top-alt:0cm;margin-right:0cm;margin-bottom:12.0pt;margin-left:36.0pt">
<b><span lang="EN-US" style="font-size:12.0pt;color:black">From: </span></b><span lang="EN-US" style="font-size:12.0pt;color:black">Asim Roy <ASIM.ROY@asu.edu><br>
<b>Date: </b>Saturday, February 5, 2022 at 11:59 PM<br>
<b>To: </b>Gary Marcus <gary.marcus@nyu.edu>, Stephen José Hanson <jose@rubic.rutgers.edu><br>
<b>Cc: </b>connectionists@mailman.srv.cs.cmu.edu <connectionists@mailman.srv.cs.cmu.edu>, Stevan Harnad <harnad@ecs.soton.ac.uk>, Francesca Rossi2 <Francesca.Rossi2@ibm.com>, Artur Garcez <arturdavilagarcez@gmail.com>, Anima Anandkumar <anima@caltech.edu>,
Luis Lamb <lamb@inf.ufrgs.br>, Gadi Singer <gadi.singer@intel.com>, Josh Tenenbaum <josh.tenenbaum@gmail.com>, AIhub <aihuborg@gmail.com><br>
<b>Subject: </b>RE: Connectionists: Stephen Hanson in conversation with Geoff Hinton<o:p></o:p></span></p>
</div>
<div>
<div style="border:solid #8D3970 1.0pt;padding:9.0pt 9.0pt 9.0pt 9.0pt">
<p class="MsoNormal" style="margin-left:36.0pt;line-height:16.5pt;background:#F7F9FA">
<strong><span lang="EN-US" style="font-size:10.5pt;font-family:"Calibri",sans-serif;color:#8D3970">CAUTION:</span></strong><span lang="EN-US" style="font-size:10.5pt;color:#8D3970"> This e-mail originated outside the University of Southampton.
<o:p></o:p></span></p>
</div>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">There was another recent attempt to take down the grandmother cell idea:<o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"><a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__eur03.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Fwww.frontiersin.org-252Farticles-252F10.3389-252Ffnins.2019.01121-252Ffull-26data-3D04-257C01-257Charnad-2540ecs.soton.ac.uk-257Cc74badd1ce3146050e7608d9e92d0c1f-257C4a5378f929f44d3ebe89669d03ada9d8-257C0-257C0-257C637797203596249360-257CUnknown-257CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0-253D-257C2000-26sdata-3D3ftfEaUyASwZpll7ko7U0C2F0Ie4VD8uBxs-252BcULKWu4-253D-26reserved-3D0&d=DwMGaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=wQR1NePCSj6dOGDD0r6B5Kn1fcNaTMg7tARe7TdEDqQ&m=Q7ibgfQ6PQZUOVD3SL8TDNTRI-P5JJCirWVDEm8Jc5FEm6qKxeYnlwII6KrH2vxQ&s=0cZt2ZzmYzGgxn5klRvH2L9up280tIod7imIVuOE6gY&e=">Frontiers
| The Value of Failure in Science: The Story of Grandmother Cells in Neuroscience | Neuroscience (frontiersin.org)</a><o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">And here’s my commentary defending grandmother cells:<o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"><a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__eur03.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Fwww.frontiersin.org-252Farticles-252F10.3389-252Ffnins.2020.00059-252Ffull-26data-3D04-257C01-257Charnad-2540ecs.soton.ac.uk-257Cc74badd1ce3146050e7608d9e92d0c1f-257C4a5378f929f44d3ebe89669d03ada9d8-257C0-257C0-257C637797203596249360-257CUnknown-257CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0-253D-257C2000-26sdata-3D3LtWHbaBaDIxgGr1a7GN1fqL0BGJ81-252Bf2cWA4UxlxRw-253D-26reserved-3D0&d=DwMGaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=wQR1NePCSj6dOGDD0r6B5Kn1fcNaTMg7tARe7TdEDqQ&m=Q7ibgfQ6PQZUOVD3SL8TDNTRI-P5JJCirWVDEm8Jc5FEm6qKxeYnlwII6KrH2vxQ&s=GwC0CbG8u8QDYfRgtlNiM65VmsvdW3nChj-1YisV7P0&e=">Frontiers
| Commentary: The Value of Failure in Science: The Story of Grandmother Cells in Neuroscience | Neuroscience (frontiersin.org)</a><o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">By the way, we have had vigorous private arguments over the years about grandmother cells and many were involved – from Jay McClelland and Christof Koch to Walter Freeman and Bernard Baars.
As far as I can tell, the brain uses abstractions at the single cell level that can be argued to be symbols. Short arguments are in the commentary, based on observations by neurophysiologists themselves.<o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">Asim Roy<o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">Professor, Information Systems<o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">Arizona State University<o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"><a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__eur03.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Furldefense.proofpoint.com-252Fv2-252Furl-253Fu-253Dhttps-2D3A-5F-5Flifeboat.com-5Fex-5Fbios.asim.roy-2526d-253DDwMFaQ-2526c-253DslrrB7dE8n7gBJbeO0g-2DIQ-2526r-253DwQR1NePCSj6dOGDD0r6B5Kn1fcNaTMg7tARe7TdEDqQ-2526m-253DwaSKY67JF57IZXg30ysFB-5FR7OG9zoQwFwxyps6FbTa1Zh5mttxRot-5Ft4N7mn68Pj-2526s-253DoDRJmXX22O8NcfqyLjyu4Ajmt8pcHWquTxYjeWahfuw-2526e-253D-26data-3D04-257C01-257Charnad-2540ecs.soton.ac.uk-257Cc74badd1ce3146050e7608d9e92d0c1f-257C4a5378f929f44d3ebe89669d03ada9d8-257C0-257C0-257C637797203596249360-257CUnknown-257CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0-253D-257C2000-26sdata-3Dk26mxcg-252B44FEq7BMEv7Rle07Wt5c79d4dlg1efyI-252FqU-253D-26reserved-3D0&d=DwMGaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=wQR1NePCSj6dOGDD0r6B5Kn1fcNaTMg7tARe7TdEDqQ&m=Q7ibgfQ6PQZUOVD3SL8TDNTRI-P5JJCirWVDEm8Jc5FEm6qKxeYnlwII6KrH2vxQ&s=o61lUyoW4TOMkwf1awtL3hvpfc1s40AbJALqBjy9dmE&e=" target="_blank">Lifeboat
Foundation Bios: Professor Asim Roy</a><o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"><a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__eur03.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Furldefense.proofpoint.com-252Fv2-252Furl-253Fu-253Dhttps-2D3A-5F-5Fisearch.asu.edu-5Fprofile-5F9973-2526d-253DDwMFaQ-2526c-253DslrrB7dE8n7gBJbeO0g-2DIQ-2526r-253DwQR1NePCSj6dOGDD0r6B5Kn1fcNaTMg7tARe7TdEDqQ-2526m-253DwaSKY67JF57IZXg30ysFB-5FR7OG9zoQwFwxyps6FbTa1Zh5mttxRot-5Ft4N7mn68Pj-2526s-253DjCesWT7oGgX76-5Fy7PFh4cCIQ-2DIfe-2DesGblJyrBiDlro-2526e-253D-26data-3D04-257C01-257Charnad-2540ecs.soton.ac.uk-257Cc74badd1ce3146050e7608d9e92d0c1f-257C4a5378f929f44d3ebe89669d03ada9d8-257C0-257C0-257C637797203596249360-257CUnknown-257CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0-253D-257C2000-26sdata-3D-252BBXqwqFM-252BTXex8q5mL-252B-252FDCI5PQFmEHHGS8fGBbpWR3s-253D-26reserved-3D0&d=DwMGaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=wQR1NePCSj6dOGDD0r6B5Kn1fcNaTMg7tARe7TdEDqQ&m=Q7ibgfQ6PQZUOVD3SL8TDNTRI-P5JJCirWVDEm8Jc5FEm6qKxeYnlwII6KrH2vxQ&s=5Qnp3b2TW61UA8nfl2CgGfwH6QeKpOguYRswFQLCJ-4&e=" target="_blank">Asim
Roy | iSearch (asu.edu)</a><o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<div>
<div style="border:none;border-top:solid #E1E1E1 1.0pt;padding:3.0pt 0cm 0cm 0cm">
<p class="MsoNormal" style="margin-left:36.0pt"><b><span lang="EN-US">From:</span></b><span lang="EN-US"> Connectionists <connectionists-bounces@mailman.srv.cs.cmu.edu>
<b>On Behalf Of </b>Gary Marcus<br>
<b>Sent:</b> Friday, February 4, 2022 12:53 PM<br>
<b>To:</b> Stephen José Hanson <jose@rubic.rutgers.edu><br>
<b>Cc:</b> connectionists@mailman.srv.cs.cmu.edu; Stevan Harnad <harnad@ecs.soton.ac.uk>; Francesca Rossi2 <Francesca.Rossi2@ibm.com>; Artur Garcez <arturdavilagarcez@gmail.com>; Anima Anandkumar <anima@caltech.edu>; Luis Lamb <lamb@inf.ufrgs.br>; Gadi Singer
<gadi.singer@intel.com>; Josh Tenenbaum <josh.tenenbaum@gmail.com>; AIhub <aihuborg@gmail.com><br>
<b>Subject:</b> Re: Connectionists: Stephen Hanson in conversation with Geoff Hinton<o:p></o:p></span></p>
</div>
</div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">Steve,<o:p></o:p></span></p>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">The phrase I always liked was “poverty of the imagination arguments”; I share your disdain for them. But that’s why I think you should be careful of any retreat into biological plausibility.
As even Jay McClelland has acknowledged, we do know that some humans some of the time manipulate symbols. So wetware-based symbols are not literally biologically impossible; the real question for cognitive neuroscience is about the scope and development of
symbols. <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">For engineering, the real question is, are they useful. Certainly for software engineering in general, they are indispensable.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">Beyond this, none of the available AI approaches map particularly neatly onto what we know about the brain, and none of what we know about the brain is understood well enough to solve AI. All
the examples you point to, for instance, are actually controversial, not decisive. As you probably know, for example, Nancy Kanwisher has a different take on domain-specificity than you do (<a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__eur03.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Furldefense.com-252Fv3-252F-5F-5Fhttps-253A-252Fweb.mit.edu-252Fbcs-252Fnklab-252F-5F-5F-253B-21-21IKRxdwAv5BmarQ-21JCW7V01kHqhTTrA2Up45kCtD9LH-5FyvmF66ZxASm8db91n2BEu2yuciVpdlGuWSU-2524-26data-3D04-257C01-257Charnad-2540ecs.soton.ac.uk-257Cc74badd1ce3146050e7608d9e92d0c1f-257C4a5378f929f44d3ebe89669d03ada9d8-257C0-257C0-257C637797203596249360-257CUnknown-257CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0-253D-257C2000-26sdata-3DbGhWJ28m6ZJAKQItBCUBRTh3EVyqTQ7zqlBEBkZtKoQ-253D-26reserved-3D0&d=DwMGaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=wQR1NePCSj6dOGDD0r6B5Kn1fcNaTMg7tARe7TdEDqQ&m=Q7ibgfQ6PQZUOVD3SL8TDNTRI-P5JJCirWVDEm8Jc5FEm6qKxeYnlwII6KrH2vxQ&s=L1_UewGLKrDr5HaWzHkKcu1lYeJ0WBbPIUcTqk28Wk8&e=">https://web.mit.edu/bcs/nklab/</a>),
with evidence of specialization early in life, and Jeff Bowers has argued that the grandmother cell hypothesis has been dismissed prematurely (<a href="https://urldefense.proofpoint.com/v2/url?u=https-3A__eur03.safelinks.protection.outlook.com_-3Furl-3Dhttps-253A-252F-252Furldefense.com-252Fv3-252F-5F-5Fhttps-253A-252Fjeffbowers.blogs.bristol.ac.uk-252Fblog-252Fgrandmother-2Dcells-252F-5F-5F-253B-21-21IKRxdwAv5BmarQ-21JCW7V01kHqhTTrA2Up45kCtD9LH-5FyvmF66ZxASm8db91n2BEu2yuciVpI7KDNxg-2524-26data-3D04-257C01-257Charnad-2540ecs.soton.ac.uk-257Cc74badd1ce3146050e7608d9e92d0c1f-257C4a5378f929f44d3ebe89669d03ada9d8-257C0-257C0-257C637797203596405566-257CUnknown-257CTWFpbGZsb3d8eyJWIjoiMC4wLjAwMDAiLCJQIjoiV2luMzIiLCJBTiI6Ik1haWwiLCJXVCI6Mn0-253D-257C2000-26sdata-3DjHBW4EkAVotSvo8GYg3yUaKHpuq9DeSIQHvO26-252F-252Bfzw-253D-26reserved-3D0&d=DwMGaQ&c=slrrB7dE8n7gBJbeO0g-IQ&r=wQR1NePCSj6dOGDD0r6B5Kn1fcNaTMg7tARe7TdEDqQ&m=Q7ibgfQ6PQZUOVD3SL8TDNTRI-P5JJCirWVDEm8Jc5FEm6qKxeYnlwII6KrH2vxQ&s=6kuxCUn6HOG-Whgq69mE3LwIpKxaMIMbfCqcxhRIci8&e=">https://jeffbowers.blogs.bristol.ac.uk/blog/grandmother-cells/</a>);
there’s also a long literature on the possible neural realization of rules, both in humans and other animals. <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">I don’t know what the right answers are there, but nor do I think that neurosymbolic systems are beholden to them anymore than CNNs are bound to whether or not the brain performs back-propagation.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">Finally, as a reminder, “Distributed” per se in not the right question; in some technical sense ASCII encodings are distributed, and about as symbolic as you can get. The proper question is
really what you do with your encodings; the neurosymbolic approach is trying to broaden the available range of options.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">Gary<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
</div>
<div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<p class="MsoNormal" style="mso-margin-top-alt:0cm;margin-right:0cm;margin-bottom:12.0pt;margin-left:36.0pt">
<span lang="EN-US">On Feb 4, 2022, at 07:04, Stephen José Hanson <<a href="mailto:jose@rubic.rutgers.edu">jose@rubic.rutgers.edu</a>> wrote:<o:p></o:p></span></p>
</blockquote>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<p style="margin-left:36.0pt"><span lang="EN-US" style="font-size:13.5pt">Well I don't like counterfactual arguments or ones that start with "It can't be done with neural networks.."--as this amounts to the old Rumelhart saw, of "proof by lack of imagination".</span><span lang="EN-US"><o:p></o:p></span></p>
<p style="margin-left:36.0pt"><span lang="EN-US" style="font-size:13.5pt">I think my position and others (I can't speak for Geoff and won't) is more of a "purist" view that brains have computationally complete representational power to do what ever is required
of human level mental processing. AI symbol systems are remote descriptions of this level of processing. Looking at 1000s of brain scans, one begins to see a pattern of interacting large and smaller scale networks, probably related to Resting state and
the Default Mode networks in some important competitive way. But what one doesn't find is modular structure (e.g. face area.. nope) or evidence of "symbols" being processed. Research on Numbers is interesting in this regard, as number representation
should provide some evidence of discrete symbol processing as would letters. But again the processing states from brain imaging more generally appear to be distributed representations of some sort.</span><span lang="EN-US"><o:p></o:p></span></p>
<p style="margin-left:36.0pt"><span lang="EN-US" style="font-size:13.5pt">One other direction has to do with prior rules that could be neurally coded and therefore provide an immediate bias in learning and thus dramatically reduce the number of examples required
for asymptotic learning. Some of this has been done with pre-training-- on let's say 1000s of videos that are relatively generic, prior to learning on a small set of videos related to a specific topic-- say two individuals playing a monopoly game. In
that case, no game-like videos were sampled in the pre-training, and the LSTM was trained to detect change point on 2 minutes of video, achieving a 97% match with human parsers. In these senses I have no problem with this type of hybrid training.</span><span lang="EN-US"><o:p></o:p></span></p>
<p style="margin-left:36.0pt"><span lang="EN-US" style="font-size:13.5pt">Steve</span><span lang="EN-US"><o:p></o:p></span></p>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">On 2/4/22 9:07 AM, Gary Marcus wrote:<o:p></o:p></span></p>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">The whole point of the neurosymbolic approach is to develop systems that can accommodate both vectors and symbols, since neither on their own seems adequate.
<o:p></o:p></span></p>
<div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">If there are arguments against trying to do that, we would be interested.<o:p></o:p></span></p>
</div>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"><br>
<br>
<br>
<o:p></o:p></span></p>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<p class="MsoNormal" style="mso-margin-top-alt:0cm;margin-right:0cm;margin-bottom:12.0pt;margin-left:36.0pt">
<span lang="EN-US">On Feb 4, 2022, at 4:17 AM, Stephen José Hanson <a href="mailto:jose@rubic.rutgers.edu">
<jose@rubic.rutgers.edu></a> wrote:<o:p></o:p></span></p>
</blockquote>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></p>
<p style="margin-left:36.0pt"><span lang="EN-US" style="font-size:13.5pt">Geoff's position is pretty clear. He said in the conversation we had and in this thread, "vectors of soft features",</span><span lang="EN-US"><o:p></o:p></span></p>
<p style="margin-left:36.0pt"><span lang="EN-US" style="font-size:13.5pt">Some of my claim is in several of the conversations with Mike Jordan and Rich Sutton, but briefly, there are a number of<br>
very large costly efforts from the 1970s and 1980s, to create, deploy and curate symbol AI systems that were massive failures. Not counterfactuals, but factuals that failed. The MCC comes to mind with Adm Bobby Inmann's national US mandate to counter the
Japanese so called"Fifth-generation AI systems" as a massive failure of symbolic AI.
</span><span lang="EN-US"><o:p></o:p></span></p>
<p style="margin-left:36.0pt"><span lang="EN-US" style="font-size:13.5pt">--------------------</span><span lang="EN-US"><o:p></o:p></span></p>
<p style="margin-left:36.0pt"><span lang="EN-US" style="font-size:13.5pt">In 1982, Japan launched its Fifth Generation Computer Systems project (FGCS), designed to develop intelligent software that would run on novel computer hardware. As the first national,
large-scale artificial intelligence (AI) research and development (R&D) project to be free from military influence and corporate profit motives, the FGCS was open, international, and oriented around public goods.</span><span lang="EN-US"><o:p></o:p></span></p>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">On 2/3/22 6:34 PM, Francesca Rossi2 wrote:<o:p></o:p></span></p>
</div>
<blockquote style="margin-top:5.0pt;margin-bottom:5.0pt">
<pre style="margin-left:36.0pt"><span lang="EN-US">Hi all.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Thanks Gary for adding me to this thread.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">I also would be interested in knowing why Steve thinks that NS AI did not work in the past, and why this is an indication that it cannot work now or in the future.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Thanks,<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Francesca.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">------------------<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Francesca Rossi<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">IBM Fellow and AI Ethics Global Leader<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">T.J. Watson Research Center, Yorktown Heights, USA<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">+1-617-3869639<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">________________________________________<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">From: Artur Garcez <a href="mailto:arturdavilagarcez@gmail.com"><arturdavilagarcez@gmail.com></a><o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Sent: Thursday, February 3, 2022 6:00 PM<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">To: Gary Marcus<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Cc: Stephen José Hanson; Geoffrey Hinton; AIhub; <a href="mailto:connectionists@mailman.srv.cs.cmu.edu">connectionists@mailman.srv.cs.cmu.edu</a>; Luis Lamb; Josh Tenenbaum; Anima Anandkumar; Francesca Rossi2; Swarat Chaudhuri; Gadi Singer<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Subject: [EXTERNAL] Re: Connectionists: Stephen Hanson in conversation with Geoff Hinton<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">It would be great to hear Geoff's account with historical reference to his 1990 edited special volume of the AI journal on connectionist symbol processing. Judging from recent reviewing for NeurIPS, ICLR, ICML but also KR, AAAI, IJCAI (traditionally ZjQcmQRYFpfptBannerStart<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">This Message Is From an External Sender<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">This message came from outside your organization.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">ZjQcmQRYFpfptBannerEnd<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">It would be great to hear Geoff's account with historical reference to his 1990 edited special volume of the AI journal on connectionist symbol processing.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Judging from recent reviewing for NeurIPS, ICLR, ICML but also KR, AAAI, IJCAI (traditionally symbolic), there is a clear resurgence of neuro-symbolic approaches.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Best wishes,<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Artur<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">On Thu, Feb 3, 2022 at 5:00 PM Gary Marcus <<a href="mailto:gary.marcus@nyu.edu">gary.marcus@nyu.edu</a><a href="mailto:gary.marcus@nyu.edu"><mailto:gary.marcus@nyu.edu></a>> wrote:<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Steve,<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">I’d love to hear you elaborate on this part,<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> Many more shoes will drop in the next few years. I for one don't believe one of those shoes will be Hybrid approaches to AI, I've seen that movie before and it didn't end well.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">I’d love your take on why you think the impetus towards hybrid models ended badly before, and why you think that the mistakes of the past can’t be corrected. Also it’ would be really instructive to compare with deep learning, which lost steam for quite some time, but reemerged much stronger than ever before. Might not the same happen with hybrid models?<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">I am cc’ing some folks (possibly not on this list) who have recently been sympathetic to hybrid models, in hopes of a rich discussion. (And, Geoff, still cc’d, I’d genuinely welcome your thoughts if you want to add them, despite our recent friction.)<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Cheers,<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Gary<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">On Feb 3, 2022, at 5:10 AM, Stephen José Hanson <<a href="mailto:jose@rubic.rutgers.edu">jose@rubic.rutgers.edu</a><a href="mailto:jose@rubic.rutgers.edu"><mailto:jose@rubic.rutgers.edu></a>> wrote:<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">I would encourage you to read the whole transcript, as you will see the discussion does intersect with a number of issues you raised in an earlier post on what is learned/represented in DLs.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Its important for those paying attention to this thread, to realize these are still very early times. Many more shoes will drop in the next few years. I for one don't believe one of those shoes will be Hybrid approaches to AI, I've seen that movie before and it didn't end well.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Best and hope you are doing well.<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US">Steve<o:p></o:p></span></pre>
<pre style="margin-left:36.0pt"><span lang="EN-US"> <o:p></o:p></span></pre>
</blockquote>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">-- <o:p></o:p></span></p>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US"><signature.png><o:p></o:p></span></p>
</div>
</div>
</div>
</blockquote>
</div>
</div>
</blockquote>
<div>
<p class="MsoNormal" style="margin-left:36.0pt"><span lang="EN-US">-- <br>
<img border="0" width="402" height="98" style="width:4.1875in;height:1.0208in" id="Picture_x0020_1" src="cid:image001.png@01D81AD9.13EC7A80"><o:p></o:p></span></p>
</div>
</div>
</blockquote>
</div>
</div>
</div>
</div></blockquote></body></html>