Check out or new pre-print on Recovery of Latent Tree Models!
arxiv.org/abs/2102.13276
English
Kluger Laboratory
6 posts
Kluger Laboratory
@KlugerLab
Department of Pathology and Applied Math Program @Yale. Developing statistical, deep learning and spectral methods for analyzing biomedical datasets.
New Haven, CT Tham gia Kasım 2020
56 Đang theo dõi63 Người theo dõi
Congratulations to lab member @GCLinderman for their new publication, demonstrating analogous paradigms for preservation of global structure in t-SNE and UMAP. Read about it here: nature.com/articles/s4158…
English
Kluger Laboratory đã retweet
A common misconception is that UMAP preserves global structure better than t-SNE. @hippopedoid and I showed that that this is only due to differences in default initialization.
Comment now published in @NatureBiotech: nature.com/articles/s4158…
Thread: twitter.com/hippopedoid/st…
Dmitry Kobak@hippopedoid
A year ago in Nature Biotechnology, Becht et al. argued that UMAP preserved global structure better than t-SNE. Now @GCLinderman and me wrote a comment saying that their results were entirely due to the different initialization choices: biorxiv.org/content/10.110…. Thread. (1/n)
English
Kluger Laboratory đã retweet
Mine and @GCLinderman's comment to the Becht et al. 2018 (@EtienneBecht) paper has finally appeared in @NatureBiotech after over a year of editorial considerations.
There is no response from the authors, so I assume we are all in agreement :-)
nature.com/articles/s4158…
English
Kluger Laboratory đã retweet
Take a look at our new preprint on Local Two-Sample Testing on Graphs, now available on @arxiv:
arxiv.org/abs/2011.03418
English