'Large sample spectral analysis of graph-based multi-manifold clustering', by Nicolas Garcia Trillos, Pengfei He, Chenghui Li.

http://jmlr.org/papers/v24/21-1254.html

#laplacians #manifolds #laplacian

'Implicit Bias of Gradient Descent for Mean Squared Error Regression with Two-Layer Wide Neural Networks', by Hui Jin, Guido Montufar.

http://jmlr.org/papers/v24/21-0832.html

#gradient #curvature #laplacian

#AMDlabnotes presents two brand new blog posts covering #GPU kernel optimization tips and tricks! 🔥

Firstly, we present a post about understanding and controlling register pressure:
https://gpuopen.com/learn/amd-lab-notes/amd-lab-notes-register-pressure-readme/?utm_source=mastodon&utm_medium=social&utm_campaign=amdlabnotes

And secondly, we present the third part of the Finite Difference Method #Laplacian series.

This blog covers even more optimizations to maximize performance on #AMD GPUs:
https://gpuopen.com/learn/amd-lab-notes/amd-lab-notes-finite-difference-docs-laplacian_part3/?utm_source=mastodon&utm_medium=social&utm_campaign=amdlabnotes (2/2)

#arxivfeed

"Large sample spectral analysis of graph-based multi-manifold clustering'
https://arxiv.org/abs/2107.13610

#data #MMC #manifolds #clustering #graph #laplacian #statistics

Why does anyone like the notation \(\Delta\) for the #Laplacian? I always thought that \(\nabla^2\) was so much more suggestive and lends itself so nicely to the equation \(\nabla^2u = \nabla \cdot (\nabla u)\).

Is it because the Laplacian is so fundamental that it gets annoying to have to always do the superscript 2?