[ExI] Neural networks score higher than humans in reading and comprehension test

[spike]

From: extropy-chat [mailto:extropy-chat-bounces at lists.extropy.org] On Behalf Of Dave Sill
Sent: Tuesday, January 16, 2018 12:24 PM
To: ExI chat list <extropy-chat at lists.extropy.org>
Subject: Re: [ExI] Neural networks score higher than humans in reading and comprehension test

 

On Tue, Jan 16, 2018 at 2:50 PM, spike <spike66 at att.net <mailto:spike66 at att.net> > wrote:

 

Dave I no longer really understand the definition of really understanding.

 

>…It's not trivial but:

 

https://en.wikipedia.org/wiki/Understanding

 

>…Is pretty good.  -Dave

 

 

Ja, cool thanks.

 

Understanding is described loosely as being enabled to make inferences.  It takes understanding in order to see the humor in satire.  Computers are terrible at that and will be for the foreseeable future.

 

The experiment with regard to reading comprehension shows that software can be given a passage in which it (very quickly) finds key phrases, rather than understanding the passage.  The problem I see with that approach: that’s how I do those tests too.  It is a skill which can be mastered, and if you do, you can really fly on those kinds of tests, go like the wind.

 

So… is there anyone else here who does reading comprehension tests that way? 

 

spike

 

 

-------------- next part --------------
An HTML attachment was scrubbed...
URL: <http://lists.extropy.org/pipermail/extropy-chat/attachments/20180116/92334a45/attachment.html>