User Portlet
Featured Contributor
| Discussions |
|---|
| (crossposted from mathematica.SE [post](https://mathematica.stackexchange.com/questions/301016/visualizing-diagrams-needed-to-compute-operatornametra3-at3)) I'm looking for help getting Mathematica code to construct diagrammatic expressions like... |
| yaroslavvb@gmail.com |
| Is there any way to use Wolfram Engine in Google Colab notebooks? https://colab.research.google.com/ I followed instructions on https://www.wolfram.com/engine/ but got stuck because Docker is not available by default (colab already runs under... |
| Btw, regarding Jax, I think of Mathematica implemented tensor differentiation natively, this would obviate the need to rely on external autodiff. The issue is that neural network training involves differentiation of expression like norm(W1 W2 W3)... |
| There's a [paper][1] which claims to implement Eigenvalue density computation for finite size Wishart matrices in Mathematica, but the authors don't include link/contact information in the paper, has anyone see this implementation by chance? For... |
| &[Wolfram Notebook][1] [1]: https://www.wolframcloud.com/obj/48f7a048-9f1b-4936-93c6-5fd45c555cd6 |
| I'm not aware of second-order methods being successful for standard neural network architectures. Shampoo exists, but it's not clear that it outperforms Adam. There's a theory that standard network architectures are already close to optimal for... |
| &[Wolfram Notebook][1] [1]: https://www.wolframcloud.com/obj/bf4ab361-2166-41a0-a183-80d83784adb0 |
| &[Wolfram Notebook][1] [1]: https://www.wolframcloud.com/obj/630b3aee-f662-4236-ad29-f71c4cfd74f4 |
| I've had about 50% rate of success at correct vectorization. One of the issues is that transformer models use the same amount of compute per prompt. Hence, harder questions need several prompts to get right. |