Event: Ideas Lab: AI-Powered Materials Discovery, Design, and Synthesis
Thursday, March 26, 2026
10 a.m. to 3 p.m.
HUB 233AB University Park Campus
Lunch is provided.
Registration deadline: Feb. 13
Recent advances in artificial intelligence, machine learning, and high-performance computing are transforming the way new materials are discovered, designed, and synthesized. Data-driven modeling, generative design, autonomous experimentation, and multi-scale simulation now make it possible to accelerate materials innovation across domains ranging from energy storage and catalysis to electronics, biomaterials, and structural materials. However, realizing this potential requires deep integration of expertise across machine learning, materials science, chemistry, physics, and advanced computing infrastructure.
This half-day Ideas Lab is designed to catalyze such integration by bringing together researchers with complementary expertise in AI/ML, materials theory and experimentation, synthesis and characterization, and scalable computing platforms. The goal is to move beyond isolated methodological advances and toward collaborative, end-to-end research projects that tightly couple data, models, experiments, and computation.
Objectives
The Ideas Lab has five primary objectives:
- Identify high-impact research opportunities at the intersection of AI, materials discovery, and synthesis that are not well addressed by current disciplinary approaches.
- Foster new interdisciplinary collaborations among researchers spanning machine learning, materials science, chemistry, physics, and advanced computing.
- Align algorithms, data, and infrastructure by explicitly considering data availability, experimental throughput, uncertainty quantification, and computational scalability.
- Develop concrete collaborative project concepts, including research questions, methodological approaches, data needs, and potential funding pathways.
- Seed sustained collaboration by enabling teams formed during the Ideas Lab to compete for follow-on seed-grant support to further develop and pilot their ideas.
Structure and Format
The Ideas Lab will be organized as an interactive, highly participatory half-day event (approximately four hours, plus lunch), structured around short provocations and focused breakout sessions rather than long formal presentations.
Opening Context (30–45 minutes)
Brief framing talks will highlight:
- The current state of AI-driven materials discovery and generative design
- Key bottlenecks in data, synthesis, characterization, and validation
- Opportunities enabled by modern computing infrastructure (HPC, cloud, experimental automation)
These talks are intended to establish a shared vocabulary and problem context rather than provide exhaustive technical reviews.
Thematic Breakout Sessions (2-2.5 hours)
Participants will self-organize into small, interdisciplinary teams around emerging themes such as:
- Data-centric materials science (curation, provenance, uncertainty, and bias)
- Foundation models and generative AI for materials design
- Closed-loop and autonomous materials synthesis
- Multi-scale modeling linking atomistic, mesoscopic, and macroscopic properties
- Integration of simulation, experiment, and real-time decision-making
Each team will be charged with developing one or more collaborative research concepts, explicitly addressing:
- Scientific and technological goals
- AI/ML methodologies and materials challenges
- Data, synthesis, and characterization workflows
- Computational and infrastructure requirements
- Validation strategies and success metrics
Synthesis, Seed-Grant Pathway, and Next Steps (45–60 minutes)
Teams will reconvene to share and refine their ideas and identify opportunities for convergence across themes. The session will conclude with an overview of the seed-grant opportunity and expectations, including scope, evaluation criteria, and timelines.
Teams formed during the Ideas Lab will be invited to submit short seed-grant proposals aimed at further developing their concepts, establishing preliminary results, and positioning projects for external funding.
Expected Outcomes
By the conclusion of the Ideas Lab, participants will have:
- Formed interdisciplinary teams around compelling research directions in AI-enabled materials discovery
- Developed well-scoped project concepts suitable for seed-grant support
- Identified shared data, experimental, and computational infrastructure needs
- Established a pipeline from ideation to pilot studies and competitive external proposals
Intended Audience
The Ideas Lab is intended for faculty, senior researchers, and advanced postdoctoral scholars with expertise in:
- Machine learning, artificial intelligence, and data science
- Materials science, chemistry, and condensed-matter physics
- Materials synthesis, characterization, and experimental automation
- High-performance, cloud, and data-intensive computing infrastructure
The format emphasizes collaborative problem formulation, cross-disciplinary exchange, and rapid team formation, with the explicit goal of launching sustained, high-impact research collaborations.