Orb latent space Tc classifier

Using what we learned when trying to use the MLFF's latent space for Tc prediction, there's a way we can simplify things for the prediction model and give it a better change of picking up on the signa

Orb latent space exploration

This is a continued deep-dive into the latent space generated by the Orb model prior to it's MLFF tasks. I have been attempting to train a model on Tc prediction using this latent space as a feature v

3D latent space exploration with Tc

Using the 256 dimensional latent space output from the Orb model, we visualize the 3DSC(MP) dataset using UMAP with direction from Tc labels. Hover a point to see Tc, formula, and Material Project identifier.

Interactive 3D latent space exploration with Tc

Using the 256 dimensional latent space output from the Orb model, we visualize the 3DSC(MP) dataset using t-SNE and UMAP. The UMAP projection has been given the target for learning a manifold that keeps similar Tc materials close together.

Orb latent space to Tc prediction

After reading the MatterSim paper, the authors proposed the idea of using the MLFF's latent space as a direct property prediction feature set. Earlier, and I had been thinking about using a VAE (or s

Notes on Orb

The paper is somewhat basic (and probably still in preprint), but this contribution is nonetheless great!

Orb Paper

Authors introduce Orb, a family of universal interatomic potentials for atomistic modeling of materials. Orb models are 3-6 times faster than existing universal potentials, stable under simulation for a range of out of distribution materials and, upon release, represented a 31% reduction in error over other methods on the Matbench Discovery benchmark. https://arxiv.org/abs/2410.22570

GHOST Meeting Notes

2025-01-03

Notes on MatterSim

Good read. Well written, very detailed and thorough. Great contribution.

MatterSim

Authors present MatterSim, a deep learning model actively learned from large-scale first-principles computations, for efficient atomistic simulations at first-principles level and accurate prediction of broad material properties across the periodic table, spanning temperatures from 0 to 5000 K and pressures up to 1000 GPa. https://arxiv.org/abs/2405.04967

Temperature ramping simulations

Sharing some things I'm learning as I work on temperature ramping simulations. The goal of these simulations is to learn how a material's lattice changes with temperature, as thermal expansion, decomp

Speed comparasion running MLIP MD on CPU vs GPU

Not exactly the most rigorous test, but this side-by-side comparison shows the difference between running MD locally (M2 Macbook Air) and on a proper server (g4dn.2xl with T4 GPU). Each log is actually 100 simulation steps too.

H2O temperature ramp

Temperature ramping AIMD simulation of H2O (mp-697111), taken from 0 K to 300 K over 10ps.

NaCl temperature ramp

Temperature ramping AIMD simulation of NaCL (mp-22851), taken from 0 K to 300 K over 10ps.

NaCl trajectories

Molecular dynamics simulation temperature ramping NaCl 3x3x3 supercell from 0 K to 300 K

H2O trajectories

Molecular dynamics simulation temperature ramping H2O 3x3x3 supercell from 0 K to 300 K

High-temperature trajectories

We had this idea before too, but cool to see Claude agrees. A lot of what we're trying to accomplish with this project requires a room temperature material. As comprehensive as Materials Project may b

Simulated temperature of H2O over 10ps from 0 K to 300 K

H2O MD simulation heating from 0 K to 300 K

Simulating ice into water

mp-697111 H2O

https://next-gen.materialsproject.org/materials/mp-697111

Calculated temperature evolution of NaCl

Langevin temperature ramp over 10ps from 0 K to 300 K on a 3x3x3 supercell of NaCl

Calculated material temperature of NaCl during temperature ramp simulation

Langevin temperature ramp over 10ps from 0 K to 300 K on a 3x3x3 supercell of NaCl

mp-22851 NaCl

https://next-gen.materialsproject.org/materials/mp-22851

Molecular dynamics heating NaCl from 0K to 300K over 10ps

Visualization created using .traj outputs of ASE MD simulation

Notes from Materials Design for New Superconductors

Some notes as I read:

What are quasiparticles?

Great video intro from PBS Space Time: https://youtu.be/le_ORQZzkmE?si=ylKXLkx5D_AfzGdE

Materials Design for New Superconductors

Since the announcement in 2011 of the Materials Genome Initiative by the Obama administration, much attention has been given to the subject of materials design to accelerate the discovery of new materials that could have technological implications. Although having its biggest impact for more applied materials like batteries, there is increasing interest in applying these ideas to predict new superconductors. This is obviously a challenge, given that superconductivity is a many body phenomenon, with whole classes of known superconductors lacking a quantitative theory. Given this caveat, various efforts to formulate materials design principles for superconductors are reviewed here, with a focus on surveying the periodic table in an attempt to identify cuprate analogues. https://arxiv.org/abs/1601.00709

vajra

Tibetan-style vajra/dorje. Based on references from https://www.himalayanart.org/search/set.cfm?setID=563. https://sketchfab.com/3d-models/vajra-1bdaddab069f424a95ad9a9a9c8c17ed

Photo-induced superconductivity

Photo-induced superconductivity is where light is used to induce superconducting-like states in materials. If we can learn more about the mechanisms behind this phenomenon, we can more intentionally d

Evidence for metastable photo-induced superconductivity in K3C60

Authors make use of a new optical device to drive metallic K3C60 with mid-infrared pulses of tunable duration, ranging between one picosecond and one nanosecond. The same superconducting-like optical properties observed over short time windows for femtosecond excitation are shown here to become metastable under sustained optical driving, with lifetimes in excess of ten nanoseconds. https://www.nature.com/articles/s41567-020-01148-1

January '25 Reading List

M3GNet seems like a pretty popular MLIP model. Depending on the pipeline we build out, we may want to increase throughput with a model that can help us with MD and electronics predictions.

Prediction of high-Tc superconductivity in ternary lanthanum borohydrides

The study of superconductivity in compressed hydrides is of great interest due to measurements of high critical temperatures (Tc) in the vicinity of room temperature, beginning with the observations of LaH10 at 170-190 GPa. However, the pressures required for synthesis of these high Tc superconducting hydrides currently remain extremely high. Here we show the investigation of crystal structures and superconductivity in the La-B-H system under pressure with particle-swarm intelligence structure searches methods in combination with first-principles calculations. https://arxiv.org/abs/2107.02553

Computational methods for predicting Tc

This post will focus on the methods available to predict/derive of a material. We want to be able to build a pipeline where we can go beyond the available (and experimental) Tc data and train a model

An invertible crystallographic representation for general inverse design of inorganic crystals with targeted properties

Realizing general inverse design could greatly accelerate the discovery of new materials with user-defined properties. However, state-of-the-art generative models tend to be limited to a specific composition or crystal structure. Herein, we present a framework capable of general inverse design (not limited to a given set of elements or crystal structures), featuring a generalized invertible representation that encodes crystals in both real and reciprocal space, and a property-structured latent space from a variational autoencoder (VAE). https://arxiv.org/abs/2005.07609

A Universal Graph Deep Learning Interatomic Potential for the Periodic Table

Here, we report a universal IAP for materials based on graph neural networks with three-body interactions (M3GNet). The M3GNet IAP was trained on the massive database of structural relaxations performed by the Materials Project over the past 10 years and has broad applications in structural relaxation, dynamic simulations and property prediction of materials across diverse chemical spaces. Chi Chen & Shyue Ping Ong https://www.nature.com/articles/s43588-022-00349-3 Preprint version from arXiv

Notes on Predicting superconducting transition temperature through advanced machine learning and innovative feature engineering

So far this is the most recent paper I've found on ML prediction of , improving on both modeling (CatBoost) and dataset compared to Stanev et al.

Predicting superconducting transition temperature through advanced machine learning and innovative feature engineering

This study employs the SuperCon dataset as the largest superconducting materials dataset. Then, we perform various data pre-processing steps to derive the clean DataG dataset, containing 13,022 compounds. In another stage of the study, we apply the novel CatBoost algorithm to predict the transition temperatures of novel superconducting materials. In addition, we developed a package called Jabir, which generates 322 atomic descriptors. We also designed an innovative hybrid method called the Soraya package to select the most critical features from the feature space. These yield R2 and RMSE values (0.952 and 6.45 K, respectively) superior to those previously reported in the literature. Finally, as a novel contribution to the field, a web application was designed for predicting and determining the Tc values of superconducting materials.

Critical temperature prediction models

Literature review of existing studies done on predicting with machine learning.

Superconductor databases

Literature review of databases with materials and . See literature review on ML models which utilize these datasets:

3DSC - a dataset of superconductors including crystal structures

Data-driven methods, in particular machine learning, can help to speed up the discovery of new materials by finding hidden patterns in existing data and using them to identify promising candidate materials. In the case of superconductors, the use of data science tools is to date slowed down by a lack of accessible data. In this work, we present a new and publicly available superconductivity dataset (‘3DSC’), featuring the critical temperature Tc of superconducting materials additionally to tested non-superconductors.

Scatter plots of Tc for superconducting materials with family-specific regression predictors

For 4000 “low-Tc” superconductors (i.e., non-cuprate and non-iron-based), Tc is plotted vs. the a) average atomic weight, b) average covalent radius, and c) average number of d) valence electrons. Having low average atomic weight and low average number of d) valence electrons are necessary (but not sufficient) conditions for achieving high Tc in this group. d) Scatter plot of Tc for all known superconducting cuprates vs. the mean number of unfilled orbitals. c), d) suggest that the values of these predictors lead to hard limits on the maximum achievable Tc

SuperCon dataset and classification model performance

a) Histogram of materials categorized by Tc (bin size is 2 K, only those with finite Tc are counted). Blue, green, and red denote low-Tc, iron-based, and cuprate superconductors, respectively. In the inset: histogram of materials categorized by ln(Tc) restricted to those with Tc > 10 K. b) Performance of different classification models as a function of the threshold temperature (Tsep) that separates materials in two classes by Tc. Performance is measured by accuracy (gray), precision (red), recall (blue), and F1 score (purple). The scores are calculated from predictions on an independent test set, i.e., one separate from the dataset used to train the model. In the inset: the dashed red curve gives the proportion of materials in the above-Tsep set. c) Accuracy, precision, recall, and F1 score as a function of the size of the training set with a fixed test set. d) Accuracy, precision, recall, and F1 as a function of the number of predictors

Notes from Machine learning modeling of superconducting critical temperature

So far a really interesting paper. Published in 2018. Adding some informal notes and interesting findings here. Finding out how much literature is based on this study.

Machine learning modeling of superconducting critical temperature

Superconductivity has been the focus of enormous research effort since its discovery more than a century ago. Yet, some features of this unique phenomenon remain poorly understood; prime among these is the connection between superconductivity and chemical/structural properties of materials. To bridge the gap, several machine learning schemes are developed herein to model the critical temperatures (Tc) of the 12,000+ known superconductors available via the SuperCon database. Materials are first divided into two classes based on their Tc values, above and below 10 K, and a classification model predicting this label is trained. The model uses coarse-grained features based only on the chemical compositions. It shows strong predictive power, with out-of-sample accuracy of about 92%. https://www.nature.com/articles/s41524-018-0085-8

A foundation model for atomistic materials chemistry

Machine-learned force fields have transformed the atomistic modeling of materials by enabling simulations of ab initio quality on unprecedented time and length scales. However, they are currently limited by: (i) the significant computational and human effort that must go into development and validation of potentials for each particular system of interest; and (ii) a general lack of transferability from one chemical system to the next. Here, using the state-of-the-art MACE architecture we introduce a single general-purpose ML model, trained on a public database of 150k inorganic crystals, that is capable of running stable molecular dynamics on molecules and materials.

AIMNet2 Paper

The 2nd generation of our atoms-in-molecules neural network potential (AIMNet2), which is applicable to species composed of up to 14 chemical elements in both neutral and charged states, making it a valuable method for modeling the majority of non-metallic compounds. Using an exhaustive dataset of 2 x 107 hybrid DFT level of theory quantum chemical calculations, AIMNet2 combines ML-parameterized short-range and physics-based long-range terms to attain generalizability that reaches from simple organics to diverse molecules with “exotic” element-organic bonding.

MLIP Models and Frameworks

https://github.com/mir-group/nequip

CHGNet as a pretrained universal neural network potential for charge-informed atomistic modelling

Here we present the Crystal Hamiltonian Graph Neural Network (CHGNet), a graph neural network-based machine-learning interatomic potential (MLIP) that models the universal potential energy surface. CHGNet is pretrained on the energies, forces, stresses and magnetic moments from the Materials Project Trajectory Dataset, which consists of over 10 years of density functional theory calculations of more than 1.5 million inorganic structures. https://www.nature.com/articles/s42256-023-00716-3

Equivariant graph neural networks for data-efficient and accurate interatomic potentials

This work presents Neural Equivariant Interatomic Potentials (NequIP), an E(3)-equivariant neural network approach for learning interatomic potentials from ab-initio calculations for molecular dynamics simulations. https://www.nature.com/articles/s41467-022-29939-5

Understanding the Dynamic Structure Factor (DSF)

The Dynamic Structure Factor (S(Q,ω)) is like a movie of how atoms move in a material. Instead of just knowing where atoms are, it tells us how they move together over time:

Material screening pipeline

Superconductivity typically emerges from strong interactions between electrons and vibrations in the crystal lattice (phonons). These interactions can lead to electron pairing, enabling resistance-fre

noise_step

Training in 1.58b With No Gradient Memory. Preprint paper by wbrickner

Funding and grant opportunities

https://www.nist.gov/chips/chips-rd-funding-opportunities