Sharing a little work I did at the beginning of this project. Here, we have a HTML file generated from Python and Plotly that displays 5000 or so magnetic materials in 3 dimensions. Each unit cell was
We're starting to bring a few of the pieces together in our permanent magnet screening pipeline. In this post we'll look at how well we are able to filter out materials from a list of ~5000 ferro/ferr
Inspired by 's Project 014, we've exposed our Curie temperature prediction model so that you can test your own materials!
Trump won the election in 2024. He's going to do a lot in 4 years. Make a lot of noise, break a lot of things, do a lot of high risk, questionable things. I think some things will work out well. I ima
As a third follow up to the AI roundtable this evening, I wanted to know about how the changing world would affect bitcoin. See the other posts in the series:
Follow up post to an experiment I was running, asking each of the big AIs what they thought about the recent developments in tariffs applied by the United States.
Doing a little experiment tonight. I've been avoiding understanding the ramifications of the recent news on tariffs and trade relationships. Going to ask all the AIs what they think. Let's see if gath
Having a little fun generating some visuals in Midjourney to represent the superconductors team. So many cool styles to test out.
We'll be back to working on superconductor soon. Come check out the work going on in #permanent-magnets in the meantime as we search for better permanent magnets.
In this post I'll share some of the work I've been doing on a Curie temperature prediction model. I finally found a decent dataset to work with. More on that here:
New MLIP model on the leaderboards! Currently #2 with an F1 score of 0.884. Congrats to the team. They provide a few pre-trained models as well as a ASE calculator for MD. Great stuff. It's a graph ne
Sharing some notes as I read this paper. I uploaded it here for reference. I came across it looking for a Curie temperature dataset and so far this has been the best I've found so far.
See the little gold check near my name? That's our Gold membership! Really excited to share this one with you guys. Gold is a $5 / month subscription that unlocks monetization features on the platform
It's starting to warm up back here in Chicago! Just a couple days ago I was complaining because it was snowing and I had to wear my parka. Then yesterday it was high 50s, sunny, and comfortable in sho
cmd+enter to send your message, save your comment, or post to a team! enter for new line, shift+enter for hard breaks. ouro is going to be silky smooth for those that love to create.
hey, you can edit comments and messages now!
Hey, I'm Matt! I'm building Ouro full-time and working on a couple materials science projects.
good time to buy bitcoin tonight
I wanted to formalize in writing the idea that I keep coming back to for end-to-end material discovery. The hardest part of this project has been actually optimizing towards materials that have some p
I started looking at MuMax3 yesterday, but realized that because of the higher level of theory, it's not really possible to directly use a CIF file in the simulations. You can approximate some of the
MuMax3 is best suited for mesoscale simulations, meaning it can model magnetization behavior at the micro- and nano-scale, which is crucial for understanding domain structures, coercivity, and thermal
thanks for using Ouro, man. I want to continue to make this place a better space for doing great work, creativity, and building your personal brand. I'd love to hear any feedback you have so far. Wha
We're starting to be able to generate some decent, better permanent magnet candidates. The next step in a end-to-end pipeline would be to quickly calculate the full set of relevant magnet properties s
made this team so we can have a space focused on magnet research. This will also help me understand how we can best share information and progress between related teams as we can continue to use the
What a beautiful place. It's lively, fully of beauty and many things to enjoy. I'm here in mid-February and the weather is perfect and the water is warm.
Extending the comparison to a different model CHGNet, this time a proper MLIP. Similar to the Orb model, this model predicts energy, force, and stress, but with the addition of the magnetic moment for
In searching for physical meaning to the latent space, I've been looking closely at how these latent features change with temperature. We started by seeing how, even from ground state, we could alread
In this next experiment, I'm going to try to build more of an intuition and physical understanding of some of our most important latent features, as they relate to importance of predicting the superc
The below animation shows a selection of important features to superconductivity and how they evolve as the materials are heated up to their critical temperature. Notice how for most features, there i
By looking at our superconducting state classifier model feature importance, we can understand what features we should be looking at. From there, we can start to study how these features change across
Great little video of a guy synthesizing YBCO in a relatively simple home lab: https://www.youtube.com/watch?v=sLFaa6RPJIU. There's the shake-and-bake method, or solid state synthesis, and a pyrophori
Over the last week or so, I've been working on making some upgrades to the superconducting state classifier model. See the first attempt here:
Hey everyone! Quick announcement for some new services added to Ouro. You can now take advantage of OpenAI and Stability AI's generative AI services on Ouro. We've worked to make these functionalities
Building on 's work with the fine-tuned MatterGen model, we evaluated 400 candidate materials for superconductivity using the Tc classification model. Prior work on fine-tuning:
In a simplified version of the full end-to-end pipeline, we attempt to use MatterGen, recently released by Microsoft, to generate novel materials and test them for superconductivity. Paper and repo
Doesn't seem to be much of an effect. Though there are more predictions made it isn't helping us find a more concrete Tc for these higher Tc materials because of all the uncertainty around the phase t
Careful evaluation of the classifier model is important so that we can truly understand the capabilities and performance of a Tc predicting model. Particularly important to us is the ability for the m
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
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
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
The paper is somewhat basic (and probably still in preprint), but this contribution is nonetheless great!
2025-01-03
Good read. Well written, very detailed and thorough. Great contribution.
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
Temperature ramping AIMD simulation of H2O (mp-697111), taken from 0 K to 300 K over 10ps.
Temperature ramping AIMD simulation of NaCL (mp-22851), taken from 0 K to 300 K over 10ps.
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
Some notes as I read:
Great video intro from PBS Space Time: https://youtu.be/le_ORQZzkmE?si=ylKXLkx5D_AfzGdE
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
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.
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
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.
Literature review of existing studies done on predicting with machine learning.
Literature review of databases with materials and . See literature review on ML models which utilize these datasets:
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.
https://github.com/mir-group/nequip
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:
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
https://www.nist.gov/chips/chips-rd-funding-opportunities