Interaction Networks for Learning about Objects, Relations and Physics · HackerPapers discussions

Interaction Networks for Learning about Objects, Relations and Physics — mentioned by stefanpie

Tue, 19 Sep 2023 16:33:24 GMT

I've work with GNNs. I have not explicitly looked into this, but in general, the dynamics of GNNs are really interesting and there are many different lenses in which one can study the behavior of GNNs and find interesting takeaways. - Explicit Message Passing: https://arxiv.org/abs/2202.11097 - Algorithmic Reasoning: https://arxiv.org/abs/2203.15544 - Diffusion: https://proceedings.mlr.press/v139/chamberlain21a.html - Spectral: https://arxiv.org/abs/2010.02863 - Sparsity: https://arxiv.org/abs/2211.15335 - Training Tricks: https://arxiv.org/abs/2108.10521 - Interaction Networks / Physical…

Paper and discussion history. A mention is not an endorsement.

Interaction Networks for Learning about Objects, Relations and Physics — mentioned by Seanny123

Sun, 05 Feb 2017 11:54:30 GMT

I'm honestly just super-pumped about any artificial intelligence system that's starting to get an intuition of physics. Google's Deep Mind put out some kind of cool stuff recentely [1], but I'm mostly just excited for anything that Ilker Yildirim [2] is doing with Joshua Tanenbaum, because it seems to triangulate more with how humans think about physics. When I was at CogSci 2016, Joshua mentioned combining this with analogical reasoning and that also sounded super cool, even though I'm not sure how to the two fit together. [1] https://arxiv.org/abs/1612.00222 [2] http://www.mit.edu/~ilkery/

Paper and discussion history. A mention is not an endorsement.