OK, let’s try to have an ML-related conversation. Any tried to train a model to solve math equations or systems of them?
How much overhead ? Any reason to use it? Original solver is not differentiable, but new one is?
E.g “Learning to Solve Hard Minimal Problems “ is faster, but less successful that Niester https://arxiv.org/abs/2112.03424, also requiring special formulation. Obviously one can use huge models like GPT3, but that’s impractical.
@ducha_aiki \
Hmm, I thought about this paper,
"On the Continuity of Rotation Representations in Neural Networks"
sanity check in screenshot
https://arxiv.org/pdf/1812.07035.pdf
@ducha_aiki
There's a closed-form soln -- w/ SVD. This paper explores that w/in NNs.
An Analysis of SVD for Deep Rotation Estimation
