Oof. I won’t be able to publish a Commonplace piece this week, partly because (or especially because) I don’t know how to do the setup for this Really Interesting Learning Theory That Is Amazing To Me But Would Put Everyone Else to Sleep If Presented Badly.
Conversation
Let’s see how people react to this teaser.
Shot: how do you learn from history, given that history doesn’t repeat itself, and that the future will always be unique and novel?
Chaser: we actually have research on this! We have bloody learning theory! You do the following …
Replying to
TLDR: case by case instead of over-generalizing, expect your priors to shift wildy, the cases/knowledge will create a network interconnected in unexpected ways, don't rely on a single model or perspective.
Reminds me a lot of your article on complex adaptive systems.
1
1
I feel like an ill-structured domain could practically be a CAS, since the 'meta' of the domain is constantly evolving.
Seems like an ill-structured domain would meet the 3 requirements in your article. No centralized control, sophisticated info processing and adaptive learning.
1