Daniel Mas Montserrat @_danielmas

Building AI at @GalateaBio @Stanford 🧮 dmasmont.github.io Joined March 2018

Tweets

167
Followers

441
Following

4K
Likes

4K

Valeriy M., PhD, MBA, CQF @predict_addict

6 months ago

How was this paper even accepted to ICLR? The commercial promoters of TabPFN are now trying to discredit one of the best open repositories, OpenML. Utterly unacceptable, how did this paper pass ethics board at ICLR?

1 3 8 3K 5

Download Image

Arturo @arturolp

a year ago

Excited to share our latest PRS work! Our @GalateaBio and @genomelink team performed a comprehensive analysis of published @PGSCatalog models along with locally trained models using LDPred2, PRS-CSx, and SNPnet, across diverse populations using @UKBIOBANK and our own data

medRxiv @medrxivpreprint

a year ago

0 3 8 2K 5

0 3 9 2K 0

medRxiv @medrxivpreprint

a year ago

Polygenic risk score portability for common diseases across genetically diverse populations medrxiv.org/cgi/content/sh… #medRxiv

0 3 8 2K 5

Yannic Kilcher 🇸🇨 @ykilcher

a year ago

No son of a construction worker is just going to randomly start doing ML research if they never hear of it and don't get told that it could be important for their future career, no matter how intelligent the kid is

16 9 279 14K 10

Daniel Mas Montserrat @_danielmas

2 years ago

Introducing "HyperFast: Instant Classification for Tabular Data" at @RealAAAI, which received the Best Paper Award at @NeurIPSConf Table rep. workshop @TrlWorkshop! We provide easy-to-use sklearn-like code: github.com/AI-sandbox/Hyp… Some insights of the work below 👇🧵(1/N)

David Bonet @d_bonet

2 years ago

1 4 22 4K 1

Download Image

1 4 15 2K 1

Daniel Mas Montserrat @_danielmas

2 years ago

This work has been led by @d_bonet with the supervision of @DocXavi and @alexGioannidis! Code available at: github.com/AI-sandbox/Hyp… #AI #AAAI #AAAI2024 #NeurIPS2023 #NeurIPS (5/5)

0 3 4 828 0

Download Image

Daniel Mas Montserrat @_danielmas

2 years ago

Hyperfast provides competitive results in several tabular classification datasets, even matching boosting-tree-based accuracies! While still far from solving tabular data classification, we believe Hyperfast provides a step forward in NN-based tabular applications! (4/N)

1 2 5 299 0

Download Image

Daniel Mas Montserrat @_danielmas

2 years ago

Hyperfast provides multiple mechanisms to scale to both large and high-dimensional datasets and can be easily applied to real-world applications! (3/N)

1 1 4 240 0

Download Image

Daniel Mas Montserrat @_danielmas

2 years ago

Hyperfast replaces the slow process of training MLPs with gradient-based methods (e.g. Adam) with a fast hypernetwork that directly predicts the weights of the MLP. The generated MLP typically matches (or even surpasses) the accuracy of those trained with gradient descent. (2/N)

1 2 6 392 2

Download Image

David Bonet @d_bonet

2 years ago

Code and pre-trained weights now available on GitHub, with a very easy-to-use scikit-learn-like interface: github.com/AI-sandbox/hyp…

0 1 3 240 0

David Bonet @d_bonet

2 years ago

We presented our paper "HyperFast: Instant Classification for Tabular Data" at #AAAI2024,which also received the Best Paper Award at the NeurIPS 2023 Table Representation Learning workshop @TrlWorkshop. Cheers to the amazing team! @_danielmas @DocXavi @alexGioannidis 🧵👇