Twitxy Saved

Kovács Dávid Péter

@davkovacs10

Machine Learning researcher
London, UK Joined September 2014
413 Following 516 Followers 325 Posts

Latest posts 5

Pinned post

Kovács Dávid Péter @davkovacs10
💥I am happy to share our work on next gen transferable ML force fields for bio-organic / molecular chemistry.

MACE-OFF23 is a series of pretrained MACE models accurately describing molecules in vacuum as well as condensed phase with a wide range of tests
arxiv.org/abs/2312.15211
arxiv.org MACE-OFF: Transferable Short Range Machine Learning Force Fields... Classical empirical force fields have dominated biomolecular simulation for over 50 years. Although widely used in drug discovery, crystal structure prediction, and biomolecular dynamics, they...
3 replies 12 reposts 70 likes 12K views View on X
Kovács Dávid Péter @davkovacs10
Check out this great summary of our foundation model for chemistry / materials science by @lars__schaaf

This thread hints at the emergent behaviour of our model with applicability well beyond the training domain of crystals. 👇👇👇
Lars Schaaf @lars__schaaf Jan 19, 2024
The first time I used this model I was pretty breath taken. The coolest things: with just a few lines, **anyone** can now run decent atomic scale simulations.

Try it out yourself! Details in 🧵
paper: arxiv.org/abs/2401.00096
View quoted post on X
1 replies 1 reposts 24 likes 2.1K views View on X
Kovács Dávid Péter @davkovacs10
Such a privilege to be part of a community that builds great science tools and has instantly 60 people testing it on 30+ scientific questions! 🤟🤟
Ilyes Batatia @IlyesBatatia Jan 2, 2024
Our large collaborative effort shows the impressive range of applicability of foundational ML potentials, developed using only open-source data and software. More than 30 applications, including MOFs, catalysis, water, and more simulated with one model.
arxiv.org/abs/2401.00096
View quoted post on X
2 replies 6 reposts 25 likes 3.5K views View on X
Kovács Dávid Péter @davkovacs10
If you had enough of the @OpenAI @sama drama consider spending some quality time on this super interesting new paper on a novel GNN architecture by @IlyesBatatia and co.

MFN-s address some of the key limitations of MPNN-s and GAT models. 👏👏👏

arxiv.org/abs/2310.10434
Ilyes Batatia @IlyesBatatia Nov 3, 2023
Excited to present Matrix Function Networks (MFNs), our new non-local GNN architecture that parameterizes many-body non-local effects using the spectrum of learnable graph operators. MFNs' structured non-locality outperforms global attention. arxiv.org/abs/2310.10434
View quoted post on X
1 replies 5 reposts 44 likes 4.7K views View on X