Do it slowly and don’t be consistent, sometimes I select the tile with 3 pixels of the thing its supposed to contain, sometimes I leave 2 or 3 tiles that clearly contain the thing, sometimes I just select a tile that doesn’t even match. Idk, it always works, I suppose the erratic behavior is what shows them I’m human or smth
@cian@cogsci
Just off the top of my head (speculation alert)
"1) assume we have a limited storage capacity so aren't good at raw memorisation"
Maybe everything gets encoded and stored, but we aren't so good at retrieval/recall of specific episodes.
Maybe that poor exact episodic retrieval is a consequence of generalisation at retrieval.
"2) even after learning something, we tend to forget details over time."
Assuming we store new episodic memories over time, the accumulation of new memories might make it harder to retrieve specific old episodes through a generalisation at retrieval mechanism. (Also, even if parts of old episodes were to randomly disappear over time, that wouldn't necessarily stop a generalisation at retrieval mechanism. A good generalisation mechanism should be able to cope with partial records of episodes.)
"Neither of these really apply to ANNs?"
Well, you could include weight decay in an ANN and there is the phenomenon of catastrophic forgetting. However, I take the relevance of most current ANNs (feedforward, weight optimising networks) to cognitive concerns with a fairly large pinch of salt.
IMO the theoretical conceptual framework of most current ANNs doesn't make contact with cognitive concerns so you can't really ask these questions of them.
So, You Are Assigned Classical Sociological Theory in the Fall…
My best advice: RUN! Of course, this is tongue-in-cheek. This is the first of two essays I am writing on teaching theory. It's been some time since I put words to paper on this (here, here, and here), and my thinking has evolved and changed. This essay is largely devoted to what I think is a nasty hangover in the guise of inertia: the insistence that we teach classical theory as required.
Hey Ya'll! I was thinking of hosting my own Mastodon Instance, but noticed that it's pretty big as in software, I'm looking for something a bit more light. I'm preferably looking for something that runs in docker, as I have that installed on my Linux book, Yes, I am using Linux as if I attempted to use Windows, MacOS I wouldn't have a great time.
Anyways, I was going to choose lemmy, but I think that's also heavy, not sure though!
GoToSocial is designed for small / single user instances. There’s more with similar goals like snac, seppo, pub, ktistec, tapir, shuttlecraft, activities.next, and microblog.pub, but I haven’t really looked into them so I’m not sure on the status of each. There’s a nice list of activitypub software at delightful fediverse apps if you want to look at more options.