@hbdft2008

Postdoc@unibas compchem, qml, conceptual DFT, electrochem

Joined July 2017
  • Tweets30
  • Following63
  • Followers52
Nice review by Bing Huang, @ferchault, and @ProfvLilienfeld on just how important DFT is to the future of AI for molecules. Still so many open questions (especially in biophysics) but quantum chemistry is definitely a fundamental part of the equation! science.org/doi/10.1126/scie…

The central role of density functional theory in the AI age

Density functional theory (DFT) plays a pivotal role in chemical and materials science because of its relatively high predictive power, applicability, versatility, and computational efficiency. We...

science.org
3
27
1
87
11,097
Our account of all the amazing contributions! Up to 30 pages of academic prose and >400 select references! Our (w @hbdft2008) review has been accepted for the upcoming thematic issue on ML at the atomic scale in @ACSChemRev #compchem arxiv.org/abs/2012.07502
2
15
78
Conceptual DFT for solid surface makes sense at finite electronic temperature only! arxiv.org/abs/2012.11259
1
2
Cool stuff from @hbdft2008 and @ProfvLilienfeld
NCCR-MARVEL@nccr_marvel14 Sep 2020
Latest from @ProfvLilienfeld lab in @NatureChemistry describes "amon”-based #MachineLearning approach that excels in modelling quantum properties of a wide range of systems. More here: nccr-marvel.ch/highlights/20… @UniBasel_en
4
12
A great article by Pascal Pernot, Bing Huang and Andreas Savin @nccr_marvel @CNRS @ScienceSorbonne @UnivParisSaclay @UniBasel - Impact of non-normal error distributions on the benchmarking and ranking of #quantum #machinelearning models - bit.ly/2EfGv4J #compchem
2
3
8
Bing Huang retweeted
Eric Topol
@EricTopol2 Sep 2019
Replying to @EricTopol @InSilicoMeds @UofT @VectorInst @biogerontology @A_Aspuru_Guzik @NatureBiotech
Here's the whole paper and a schematic of what they achieved w/ a GAN #AI strategy. Designed, synthesized and validated (both in vivo and in vitro) a new drug, from scratch.
2
50
9
115
Advancing Drug Discovery via Artificial Intelligence cell.com/trends/pharmacologi…
Revealing Energetics of Surface Oxygen Redox from Kinetic Fingerprint in Oxygen Electrocatalysis pubs.acs.org/doi/abs/10.1021…

Revealing Energetics of Surface Oxygen Redox from Kinetic Fingerprint in Oxygen Electrocatalysis

Abstract. The key step for rational catalyst design in heterogeneous electrocatalysis is to reveal the distinctive energy profile of redox reactions of a c

pubs.acs.org
...the finding appears to support the emerging hypothesis that parts of the brain operate at a “critical point”... quantamagazine.org/deep-lear…

A Common Logic to Seeing Cats and Cosmos | Quanta Magazine

New research suggests physicists, computers and brains employ the same procedure to tease out important features from among other irrelevant bits of data.

quantamagazine.org
Approaching coupled cluster accuracy with a general-purpose neural network potential through transfer learning disq.us/t/3g6aaxz
Construction of Machine Learned Force Fields with Quantum Chemical Accuracy: Applications and Chemical Insights arxiv.org/abs/1909.08565 #compchem
6
12
The impressive ML prediction of electron density on polypeptides (and more) is now out in the 2019 Chemical Science Hot Article Collection: doi.org/10.1039/C9SC02696G We hope you will like it as much as we do.
COSMO Lab 🇨🇭🇪🇺@lab_COSMO2 Jul 2019
Replying to @lab_COSMO
We start with a *much* improved electron charge learning effort with the @corminboeuf_lab crew, within the @nccr_marvel. Spoiler alert: we can predict charge density on polypeptides. Find out more on chemrxiv.org/ndownloader/art…
2
9
39
Great collaboration with @keeeto2000 @danddavis @lonepair Here is our @nature review and perspective on #MachineLearning learning in chemical sciences! nature.com/articles/s41586-0… #compchem
10
39
Large-Scale Screening of Zeolite Structures for CO2 Membrane Separations pubs.acs.org/doi/10.1021/ja4…

Large-Scale Screening of Zeolite Structures for CO2 Membrane Separations

Abstract. We have conducted large-scale screening of zeolite materials for CO2/CH4 and CO2/N2 membrane separation applications using the free energy landsc

pubs.acs.org
[1903.07930] Recycling knowledge to explore materials: a connector theory approach arxiv.org/abs/1903.07930
First @ChemRxiv paper in our group! Get in touch if this work is relevant/interesting to you.
ChemRxiv@ChemRxiv2 Apr 2019
Towards a Design of Active Oxygen Evolution Catalysts: Insights from Automated Density Functional Theory Calculations and Machine Learning by Zachary Ulissi & co-workers bit.ly/2YCxN5Q #compchem #AIChemPreprints
4
24
Today I learned what a lustrum was after seeing it in the opening line of the abstract in pubs.acs.org/doi/abs/10.1021….

Outlining the Scaling-Based and Scaling-Free Optimization of Electrocatalysts

Abstract. For nearly 15 years, significant efforts have been directed toward the computational design of electrocatalysts for a variety of important reacti

pubs.acs.org
1
1
4
Bing Huang retweeted
Simon Olsson@smnlssn15 Apr 2019
Our new paper on building coarse-grained force-fields using neural networks trained using a force matching principle and physically motivated priors is online now: pubs.acs.org/doi/abs/10.1021… #compchem

Machine Learning of Coarse-Grained Molecular Dynamics Force Fields

Abstract. Atomistic or ab initio molecular dynamics simulations are widely used to predict thermodynamics and kinetics and relate them to molecular structu

pubs.acs.org
30
2
108
New blog post: "A Recipe for Training Neural Networks" karpathy.github.io/2019/04/2… a collection of attempted advice for training neural nets with a focus on how to structure that process over time
72
1,622
176
4,693
Inclusion of learning curves combined with training and prediction timings should become the new standard in #quantummachinelearning #quantum #machinelearning #COMPCHEM
Thomas Miller@tfmiller36 Sep 2019
Replying to @tfmiller3 @ProfvLilienfeld @MicheleCeriotti @A_Aspuru_Guzik @adrian_roitberg @olexandr @itamblyn @corminboeuf_lab @andrewwhite01 @waller_lab @teodorolaino @PaesaniLab @jppiquem
Timings also provided (Fig 8 of arXiv:1909.02041).
4
12
Load more