Call for Contributions
The development of new materials and production processes and the customization of existing ones is increasingly driven by AI, in particular Bayesian optimization and surrogate modeling. In many cases, materials science has relied on compute-intensive simulations to evaluate the properties of proposed designs, or the effect a change might have. Such simulations do not scale to the vast design spaces that materials scientists explore. Machine learning provides an alternative: properties are approximated through the predictions of surrogate models rather than computed by simulations, orders of magnitude faster.
Both AI and materials science are working on conceptually similar problems — how to efficiently identify the best design choices, be that for a machine learning pipeline or a new material. Yet, there is little collaboration between the communities. The purpose of this Bridge is to bring the communities closer together, facilitate cross-disciplinary collaborations, identify common problems, and develop plans for tackling them.
We solicit poster submissions that present novel applications, novel algorithms, or pose challenges at the intersection of AI and materials science, in the widest sense. Whether it’s a mature system or only an idea, we welcome your submissions. Areas of interest include, but are not limited to: Bayesian optimization, reinforcement learning, surrogate modeling, neural network approaches and their applications to design new materials and production processes, optimize existing materials and production processes, characterize or test materials, and monitor the performance of materials. Please feel free to contact the organizers informally for any questions.
Posters will undergo a light review by the organizers for suitability for the Bridge. Submissions are due November 18, notifications will be sent out December 2. Please submit a PDF version of your poster on Easychair at https://easychair.org/conferences/?conf=aimat23.
Areas
Machine learning and Bayesian optimization in AI, Materials science with individual disciplines (e.g. Physics, Chemistry, Chemical Engineering, Mechanical Engineering).
Outline
What are the obstacles to applying approaches developed in AI in materials science?
What problems in materials science are not addressed by state-of-the-art AI?
What common problems do both fields tackle, and how can we join efforts?
Topics to be covered include recent advances and current problems in AI for materials science, what AI could do to facilitate increased adoption of AI methods in materials science, and the challenges this particular application domain poses to AI.
Format
Participants submit posters highlighting research at the intersection of AI in materials science (e.g., novel applications, novel algorithms targeting this space), which will undergo a light review for suitability by the organizers. Posters will be available on the Bridge website; we will endeavor to record the invited talks, tutorial, and demonstration and make them available on the Bridge website as well.
The proposed schedule allocates most of the afternoon to interactive sessions to facilitate community building and encourage collaborations that last beyond AAAI. We anticipate that the breakout sessions will crystallize common interests and teams for collaborations, aided by the software demonstration in the morning, which will make the abstract concepts more concrete and tangible.
We will organize a panel discussion at the main AAAI conference to communicate the state of the art, challenges, and opportunities for AI in materials science.
Schedule
9.00 – 9.15: Welcome, introductions, logistics, etc.
9.15 – 11.00: Tutorial on Bayesian optimization (targeted to graduate students and researchers in both areas)
11.00 – 11.15: Break
11.15 – 12.15: Demonstration of software for Bayesian optimization and applications in materials science (code demo)
12.15 – 13.30: Lunch break
13.30 – 14.00: Bayesian Optimization for Peptide Design: Controlling Ice, Binding Metals, and Making Proteins Glow, Peter Frazier
14.00 – 14.15: Break
14.15 – 15.15: Poster session
15.15 – 15.45: Tough Materials by Design: Autonomous Experimentation for Extreme Mechanics, Keith Brown
15.45 – 16.00: Break
16.00 – 17.00: Breakout sessions to plan future activities, identify directions to pursue, discuss open problems
17.00 – 17.30: Reports from breakout sessions and closing
Presented Posters
Bayesian Optimization of Photonic Curing Perovskite Solar Cells (poster) by Weijie Xu, Anusha Srivastava and Julia Hsu
Bayesian Optimization with Partial Evaluations for Materials Design (poster) by Poompol Buathong, Jiayue Wan and Peter Frazier
Machine Learning Approaches in Agile Manufacturing with Recycled Materials for Sustainability (poster) by Aparna Varde and Jianyu Liang
Optimization of laser-induced graphene manufacturing (poster) by Hud Wahab, Lars Kotthoff and Patrick Johnson
Scaling Up AI-driven Scientific Discovery via Embedding Physics Modeling into End-to-end Learning and Harnessing Random Projection (poster) by Yexiang Xue, Md Nasim, Xinghang Zhang and Anter El-Azab
Organizers
Roman Garnett
Washington University in St Louis
garnett@wustl.edu
Patrick Johnson
University of Wyoming
pjohns27@uwyo.edu
Jessica Koehne
NASA
jessica.e.koehne@nasa.gov
Lars Kotthoff
University of Wyoming
larsko@uwyo.edu