The webinar “AI-driven Search for Functional Materials: Symbolic Inference in Catalysis” hosted by Materials Square, features Dr. Luca M. Ghiringhelli from Humboldt University of Berlin. The event highlights the use of AI and symbolic inference to accelerate the discovery and optimization of materials, with a particular focus on catalysis.
Key points discussed in the webinar include:
-
AI in Materials Discovery: Prof. Ghiringhelli explores how AI can extract hidden patterns from experimental and theoretical data, helping to predict material properties and behaviors effectively. This is crucial for constructing material property maps that guide researchers in identifying optimal materials for specific applications.
-
Symbolic Inference: The webinar emphasizes the role of symbolic inference in creating predictive models and descriptors that define distinct regions in material property maps. This method is particularly beneficial for working with limited data sets (“small-data”).
-
FAIRmat Consortium and NOMAD AI Toolkit: Prof. Ghiringhelli introduces the FAIRmat consortium and the NOMAD AI toolkit, platforms designed for sharing and publishing AI tools and workflows. These resources aim to facilitate collaborative research and enhance the accessibility of AI methodologies in material science.
-
Practical Applications: The webinar includes case studies and practical examples, showcasing how these AI-driven approaches can be applied to predict the catalytic properties of materials. This demonstrates the real-world impact and potential of integrating AI into materials research.
The event aims to provide insights into the latest developments in AI-driven materials science, offering valuable knowledge for researchers and professionals in the field.
For more information, you can watch the full webinar recording under the link below:
