View Single Post
Old 11-27-2022, 04:10 PM   #77
andyh2000
Avid reader
andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.andyh2000 ought to be getting tired of karma fortunes by now.
 
andyh2000's Avatar
 
Posts: 884
Karma: 6399168
Join Date: Apr 2009
Location: UK
Device: Samsung Galaxy Z Flip 4 / Kindle Paperwhite / TCL Nxtpaper 14
Quote:
Originally Posted by Quoth View Post
...
Computer Neural Networks are a kind of data-flow pattern matching (not recognition) and are nothing to do with biological neural systems. Marketing lingo.
...
So you're saying that the neural networks in certain kinds of machine learning applications aren't networks of nodes with connections between them, and the connections don't get reinforced by 'right' answers during a training phase? Much like we currently think neurons and synapses in brains work? I must be reading this article completely incorrectly then: https://en.wikipedia.org/wiki/Artificial_neural_network

How else do our brains work other than by pattern matching against stuff stored during a (life-long, ongoing) learning phase?

Edited to add - also who cares if machines don't do things the same way we do. Aeroplanes don't flap their wings but we still say they fly.

Andrew
andyh2000 is offline   Reply With Quote