Skip to main content
U.S. flag

An official website of the United States government

Official websites use .gov
A .gov website belongs to an official government organization in the United States.

Secure .gov websites use HTTPS
A lock ( ) or https:// means you’ve safely connected to the .gov website. Share sensitive information only on official, secure websites.

2024 Artificial Intelligence for Materials Science (AIMS) Workshop

AIMS 2024 Workshop
Credit: Crissy Robinson

As a part of the JARVIS workshop series, NIST sponsored the 5th Artificial Intelligence for Materials Science (AIMS) workshop. The workshop was held in-person at the National Cybersecurity Center of Excellence (NCCoE), located at 9700 Great Seneca Highway, Rockville, MD 20850, from July 17 - 18, 2024. 

The Materials Genome Initiative (MGI) promises to expedite materials discovery through high-throughput computation and high-throughput experiments. The application of artificial-intelligence (AI) tools such as machine-learningdeep-learning and various optimization techniques is critical to achieving such a goal.

Some of the key research areas for materials AI include: developing well-curated and diverse datasets, choosing effective representations for materials, inverse materials design, integrating autonomous experiments and theory, merging physics-based models with AI models, and choosing appropriate algorithms/work-flows. Lastly, uncertainty quantification in AI-based predictions for material properties and issues related to building infrastructure for disseminating AI knowledge are of immense importance for making AI- based materials investigation successful. This workshop is intended to cover all the above-mentioned challenges. To make the workshop as effective as possible we plan to largely but not exclusively focus on inorganic solid-state materials.
 

Topics addressed in this workshop will include (but not be limited to):

1)  Datasets and tools for employing AI for materials

2) Integrating experiments with AI techniques

3) Graph neural networks for materials

4) Comparison of AI techniques for materials

5) Challenges of applying AI to materials

6) Uncertainty quantification and building trust in AI predictions

7)  Generative modeling

8) Using AI to develop classical force-fields

9) Natural language processing/Large language models


CLICK HERE to check out our GitHub.


CONFIRMED SPEAKERS

  • Anouar Benali (Argonne National Lab)
  • Ale Strachan (Purdue)
  • Eddie Kim (Cohere)
  • Tian Xie (Microsoft)
  • Christopher Sutton (South Carolina)
  • Hongliang Xin (Virginia Tech)
  • Nicola Marzari (EPFL)
  • Sergei Kalinin (University of TN, Knoxville, PNNL)
  • Chris Stiles (JHUAPL)
  • Yongqiang Cheng (ORNL)
  • Mathew Cherukara (Argonne National Lab)
  • Olga Wodo (University at Buffalo)
  • Ming Hu (South Carolina)
  • Keqiang Yan (TAMU)
  • P. Ganesh (ORNL)
  • Rama Vasudevan (ORNL)
  • Guido von Rudorff (University of Kassel)
  • Maria Chan (Argonne National Lab)
  • Vidushi Sharma (IBM)
  • Debra Audus (NIST)
  • Dilpuneet Aidhy (Clemson)
  • Michael Waters (Northwestern)
  • Timur Bazhirov (Mat3ra)
  • Anuroop Sriram (Meta)


co-organizers

  • Daniel Wines
  • Kamal Choudhary
  • Francesca Tavazza
  • Kevin Garrity
  • Brian DeCost
  • Howie Joress
  • Austin McDannald

Artificial Intelligence for Materials Science (AIMS) Workshop Agenda

DATE: July 17, 2024

Start Time 

End 
Time 

Speaker(s)  

Session Name/Information 

9:009:10Jim WarrenOpening remarks by Jim Warren
9:109:25Kamal ChoudharyOverview and Logistics
9:2511:45

Nicola Marzari (9:25-9:45)

  • Machine learning electrochemistry

P. Ganesh, Abdulgani Annaberdiyev (9:45-10:05)

  • Predicting Quantum Monte Carlo Charge Densities using Graph Neural Networks

Anouar Benali (10:05-10:25)

  • Increasing AI/ML Predictions Through DMC-enhanced Delta Learning

Break (10:25-10:45)

Ming Hu (10:45-11:05)

  • Unleashing the Power of Artificial Intelligence for Phonon Thermal Transport

Christopher Sutton (11:05-11:25)

  • Machine learning models for accelerating materials discovery

Hongliang Xin (11:25-11:45)

  • Accelerating Scientific Discovery in Catalysis with Artificial Intelligence

Invited Session I

Chair: Daniel Wines

12:001:00Lunch 
1:003:20

Sergei Kalinin (1:00-1:20)

  • Integrating Autonomous Systems for Advanced Material Discovery: Bridging Experiments and Theory Through Optimized Rewards

Mathew Cherukara (1:20-1:40)

  • HPC+AI-enabled Materials Characterization and Experimental Automation

Chris Stiles (1:40-2:00)

  • Targeted AI-Driven Materials Discovery

Break (2:00-2:20)

Rama Vasudevan (2:20-2:40)

  • Algorithms and opportunities for self-driving laboratories: model-based control, physics discovery, and co-navigating theory and experiments

Maria Chan (2:40-3:00)

  • Theory-informed AI/ML for materials characterization

Yongqiang Cheng (3:00-3:20)

  • Data-driven approaches to lattice dynamics and vibrational spectroscopy

Invited Session II

Chair: Howie Joress

3:204:00Sergei Kalinin, Hongliang Xin, Chris Stiles, Rama Vasudevan, Maria Chan, Vidushi Sharma, Timur Bazhirov

Panel Discussion

Moderator: Brian DeCost

4:005:30Poster Session 

DATE: July 18, 2024

Start Time 

End 
Time 

Speaker(s)  

Session Name/Information 

8:4511:45

Timur Bazhirov (8:45-9:05)

  • Data Standards: the key enabler of AI-driven materials science at the nanoscale

Vidushi Sharma (9:05-9:25)

  • Chemical Foundation Models for Complex Materials

Eddie Kim (9:25-9:45)

  • A Practical Guide to Building with LLMs

Anuroop Sriam (9:45-10:05)

  • Beyond Experimental Structures: Advancing Materials Discovery with Generative AI

Break (10:05-10:25)

Tian Xie (10:25-10:45, Virtual)

  • Accelerating materials design with AI emulators and generators

Ale Strachan (10:45-11:05)

  • Combining machine-learning, physics, and infrastructure to accelerate materials research

Debra Audus (11:05-11:25)

  • Improving machine learning with polymer physics

Dilpuneet Aidhy (11:25-11:45)

  • Integrated Data Science and Computational Materials Science in Complex Materials

Invited Session III

Chair: Francesca Tavazza

12:001:00Lunch 
1:002:20

Michael Waters (1:00-1:20)

  • Sampling Strategies for Robust MLIPs

Guido von Rudorff (1:20-1:40)

  • Unbiased Sampling of Chemical Space

Olga Wodo (1:40-2:00)

  • Data-driven microstructure-property mapping: the importance of microstructure representation 

Keqiang Yang (2:00-2:20)

  • Artificial Intelligence for Materials Geometric Representation Learning and High Tensor Order Property Predictions

Invited Session IV

Chair: Kevin Garrity

2:304:30
  1. Peter Bajcsy (2:30-2:50)
  2. Austin McDannald (2:50-3:35)
  3. Brian DeCost/Daniel Wines/Kamal Choudhary (3:35-4:30) 

Hands On Session

  1. NN Calculator Tutorial
  2. Active Learning/Gaussian Processes
  3. ALIGNN and ALIGNN-FF 
  4. Atom-GPT

*Visitor Access Requirement: 

For Non-US Citizens:  Please have your valid passport for photo identification. 

For US Permanent Residents: Please have your green card for photo identification. 

For US Citizens: Please have your state-issued driver's license. Regarding Real-ID requirements, all states are in compliance or have an extension through May 2025. 

NIST also accepts other forms of federally issued identification in lieu of a state-issued driver's license, such as a valid passport, passport card, DOD's Common Access Card (CAC), Veterans ID, Federal Agency HSPD-12 IDs, and Military Dependents ID. 

Created April 8, 2024, Updated July 30, 2024