Looks cool! Heads up, my collaborators and I derived the state-action Linearly Solvable MDP a while back. You might be interested arxiv.org/pdf/2007.02527
25.05.2025 15:51 โ ๐ 3 ๐ 0 ๐ฌ 1 ๐ 0@noreward4u.bsky.social
Reward-Free Model-based Maximalist. High-dimensional Empowerment. Self-Preserving Autonomous Agents. Theories of intelligence grounded in compositional control.
Looks cool! Heads up, my collaborators and I derived the state-action Linearly Solvable MDP a while back. You might be interested arxiv.org/pdf/2007.02527
25.05.2025 15:51 โ ๐ 3 ๐ 0 ๐ฌ 1 ๐ 0First time seeing a wild giraffe. Just chillinโ on the side of the road.
28.03.2025 14:04 โ ๐ 3 ๐ 0 ๐ฌ 0 ๐ 0Thesis: drive.google.com/file/d/1peG7...
24.03.2025 21:17 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0Stoffel lives at an animal rehabilitation center near Kruger National Park and is an expert escape artist. But he is 26 now (they only live an average of 8 years in the wild) so he spends most days snuggling with his girlfriend Hammie. BBC show: m.youtube.com/watch?v=c36U...
24.03.2025 21:17 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0They say donโt meet your heroes, but I traveled to South Africa and met mine and it was worth it. Stoffel the Honey Badger became a major inspiration for my PhD thesis when my advisor showed our lab a BBC show on clever animals who can solve long horizon tasks, presumably for abstract reasons.
24.03.2025 21:17 โ ๐ 3 ๐ 1 ๐ฌ 1 ๐ 0And I've always wondered how this works with his Constructor theory.
14.03.2025 19:26 โ ๐ 1 ๐ 0 ๐ฌ 0 ๐ 0By the way, while Deutsch doesn't have a deeply rigorous decision theory to match his views, I did once hear him say (on a podcast I can't seem to find) that he regards value as equivalent to the space of possible transformations one can make, which is to a close approximation what empowerment is.
14.03.2025 19:23 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0I've been thinking recently about Bostrom's notion of instrumental convergence, and what would entail instrumental divergence. There's an obvious sense in which infinitely large time horizons and infinitely small relevant probabilities play a role in washing out potential differences.
13.03.2025 10:09 โ ๐ 5 ๐ 1 ๐ฌ 1 ๐ 0Deutsch's emphasis on universal explainers is a better (though incomplete) alternative and has nothing to do with emulation (he never talks about the normative part of why one should want to explain something)
13.03.2025 12:13 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0Yeah, their emphasis on emulation is frustrating (for somewhat similar reasons to the recent Jaeger/Vervaeke paper). AI-by-learning being intractable is not interesting. It doesn't imply anything about the intractability of creating generally intelligent systems.
13.03.2025 12:13 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0Related:
bsky.app/profile/nore...
Unfortunately for SlipFrosty, a theory of instrumental intelligence is inseparable from a theory of normative intelligence. Abolish the value function!
06.01.2025 17:53 โ ๐ 7 ๐ 2 ๐ฌ 1 ๐ 0My favorite interview from the past year, of philosopher Pete Wolfendale. Recommended to anyone interested in AI, the relationship between value, aesthetics and ethics, or anyone who wants a reason to abandon "rationality as Bayes + Utility".
www.youtube.com/watch?v=0xMc...
IMO, A problem with RL is that, in sparse-reward problems, value functions donโt have a general decomposition over high-dimensional transition kernels so people are trying to learn neural-net approximations to difficult-to-generalize functions from a lot of experience.
Fun ep.
Exactly :)
17.12.2024 16:36 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0@denizrudin.bsky.social Deniz, did your grandma and grandpa call you baby Rudin?
08.12.2024 14:47 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0Love that Fog Lake song.
07.12.2024 15:47 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Stellar new work lead by the inimitable James Whittington in Neuron that develops a new theory unifying episodic and working memory and explains diverse hippocampal and prefrontal data: www.cell.com/neuron/fullt... w/Will Dorrell, @behrenstimb.bsky.social Mohamedy El Gaby
06.12.2024 18:47 โ ๐ 110 ๐ 30 ๐ฌ 2 ๐ 1Starter packs are making it easier to keep constructing niches, so I made this one for people broadly interested in intrinsic motivation, with some focus on RL, neuroscience and cognitive science. Havenโt found many people on Bluesky that I wanted to add here, so help me out! go.bsky.app/TPNrnpE
26.11.2024 03:19 โ ๐ 7 ๐ 4 ๐ฌ 3 ๐ 1I wish we could turn some of the starter-packs into a custom feed rather than following everyone.
25.11.2024 17:37 โ ๐ 11 ๐ 1 ๐ฌ 6 ๐ 0@bsky.app Please consider this!
25.11.2024 17:45 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0I wish we could turn some of the starter-packs into a custom feed rather than following everyone.
25.11.2024 17:37 โ ๐ 11 ๐ 1 ๐ฌ 6 ๐ 0@saxelab.bsky.social
25.11.2024 16:49 โ ๐ 3 ๐ 0 ๐ฌ 2 ๐ 0Which ones do you have?
25.11.2024 00:19 โ ๐ 0 ๐ 0 ๐ฌ 0 ๐ 0Probably this one:
open.spotify.com/album/0Dg1Kw...
and
open.spotify.com/album/1ehwJG...
And you're all three??
22.11.2024 21:55 โ ๐ 0 ๐ 0 ๐ฌ 1 ๐ 0Thanks!
19.11.2024 12:20 โ ๐ 2 ๐ 0 ๐ฌ 0 ๐ 0I know the name, never looked into his work. Is there something I should check out?
19.11.2024 12:10 โ ๐ 2 ๐ 0 ๐ฌ 1 ๐ 0I'm with you haha, I didn't feel anything about the latest movie, other than "this orange tone is nice".
18.11.2024 21:13 โ ๐ 1 ๐ 0 ๐ฌ 1 ๐ 0