Want to take part in these discussions? Sign in if you have an account, or apply for one below
Vanilla 1.1.10 is a product of Lussumo. More Information: Documentation, Community Support.
1 to 3 of 3
There are 2 interesting articles this year: https://arxiv.org/abs/2104.03902 - about autodidactive universe that maps old-school gauge theories to old-school neural networks https://arxiv.org/abs/2106.14587 - about utmost generalization of deep neural networks (and their learning and the emergence of semantics in them) in the terms of category theory
Since the time of their appearance I have had the silent dream to make a generalization of “autodidactic universe” to the level of M-theory and topos-stacks preprint may be the right tool to do this. Is this the prospective research theme? Anyone has similar ideas?
This is just shy reminder that this forum still hosts interesting (and still open and ongoing, although without too heavy traffic) discussion about topos and stacks DNN article https://nforum.ncatlab.org/discussion/13133/understanding-preprint-topos-and-stacks-of-deep-neural-networks and that this discussion should be resolved before trying to apply topos-and-stacks-DNN to the M-theory for autodidactic (multi)uni-verse. Of course, more earthly applications are also possible.
This is just shy reminder that this forum still hosts
No need for such reminders. Already that thread “Understanding preprint…” is borderline. There is hope that user “rongmin”with their expertise turns it into something useful, but maybe that requires you to pick up and follow-up on the hints that rongmin is kindly providing for you.
Hi Tomr,
https://arxiv.org/abs/2104.03902 - about autodidactive universe that maps old-school gauge theories to old-school neural networks
Thanks for bringing this to my attention, but I’m not sure if this is the appropriate forum to discuss that paper. I find it amusing, though, that Lee Smolin is now interested in the landscape.
I am unfortunately in the midst of helping out with some mathematical “firefighting”, metaphorically speaking, so I can’t yet answer your questions for this week. In the meantime, as Urs has kindly pointed out, please review the suggestions I’ve given, which were made in light of your claim that you’re a beginner in category theory. I want to be supportive of your learning journey and those suggestions were tailored to help a beginner, so please follow up on them.
1 to 3 of 3