Jump to content

Talk:Connectionism

Page contents not supported in other languages.
From Wikipedia, the free encyclopedia


Wiki Education Foundation-supported course assignment

[edit]

This article was the subject of a Wiki Education Foundation-supported course assignment, between 20 August 2019 and 7 December 2019. Further details are available on the course page. Student editor(s): CThomC.

Above undated message substituted from Template:Dashboard.wikiedu.org assignment by PrimeBOT (talk) 18:20, 16 January 2022 (UTC)[reply]

"Connectionism" was coined in the 1980s and referred to more than just ANNs / Perceptrons

[edit]

This is just from my own memory as an undergraduate who read Parallel Distributed Processing when it first came out.

"Connectionism" was a very loose term when it first started being used, and was also used to refer semantic networks, entity-relationship models, markov decision processes, and a number of other things. "Connectionism" was more of a philosophy or criticism term, and not anything like a unified research program. It was unclear at the time what it all meant or if it mattered. That was my impression.

I removed the attribution to Thorndike, as his theory of "connectionism" isn't remotely related.

I'm sorry but I don't have sources for this, but I think Nils John Nilsson's history of AI might shed some light. He was right in the middle of the argument and sympathetic to both sides. CharlesTGillingham (talk) 21:03, 20 August 2024 (UTC)[reply]

Since machine learning imitates human learning, the reference may be appropriate after all: Edward Thorndike (1931) Human Learning, page 122 — Rgdboer (talk) 22:47, 20 August 2024 (UTC)[reply]