ADVANCED MATERIALS & PROCESSES | MARCH 2026 5 RESEARCH TRACKS ADVANCED MATERIALS COURTESY OF AI Researchers at Penn State developed a new AI framework with potential implications for fields ranging from Alzheimer’s disease research to advanced materials design. The approach, named ZENN, teaches AI models to recognize and adapt to hidden differences in data quality rather than ignoring them. ZENN, short for Zentropy- Embedded Neural Networks, was developed by Zi-Kui Liu, FASM, professor of materials science and engineering, and his colleagues Shun Wang, Wenrui Hao, and Shunli Shang. Zentropy is Liu’s advanced theory of entropy, which posits that systems tend to move towards disorder in the absence of energy to maintain order. This deeper theory of entropy integrates quantum mechanics, thermodynamics, and statistical mechanics into a cohesive predictive model. The researchers used this approach to develop their framework, embedding principles from thermodynamics directly into neural networks to allow models to distinguish meaningful signals from uncertainty and noise. ZENN takes an approach inspired by thermodynamics by breaking down data properties into two parts. One, called energy, captures the meaningful patterns or signals in the data. The other, called intrinsic entropy, captures the noise, uncertainty, or disorder in the measurements. The model also uses a “temperature” parameter that can be tuned, which helps it recognize hidden differences between datasets, such as whether the data comes from precise simulations or noisier experiments. This allows ZENN to focus on the true signal while accounting for varying data quality. In materials science, ZENN could help bridge the gap between idealized computer simulations and real-world experiments, according to Liu. By learning from both, the framework could guide the design of materials that are not only theoretically promising but also manufacturable, with potential applications ranging from medical implants for bone repair to advanced data platforms such as ULTERA, a system that manages and analyzes large, complex datasets. psu.edu. AI ASSISTANT FOR ENERGY MATERIALS DISCOVERY Spearheaded by the DOE’s Lawrence Berkeley National Laboratory, a new multi- institutional project will use artificial intelligence (AI) and supercomputers to speed discovery of materials for batteries, semiconductors, and other energy technologies. The project, FORUM-AI (Foundation Models Orchestrating Reasoning Agents to Uncover Materials Advances and Insights), supports the Genesis Mission, a new national initiative led by the DOE to advance AI and accelerate discovery, providing solutions for challenges in science, energy, and national security. “FORUM-AI aims to be the first fullstack, agentic AI system for materials science research and discovery,” says principal investigator Anubhav Jain, a staff scientist at Berkeley Lab leading the project. “It will help scientists at every step of energy materials research, from hypothesis generation and computer simulations to laboratory experiments and analysis.” The initiative is a collaboration between Berkeley Lab, Oak Ridge National Laboratory, Argonne National Laboratory, the Massachusetts Institute of Technology, and The Ohio State University with a goal to develop an open-source, general- purpose AI platform for research in materials and the physical sciences. lbl.gov. Illustration of ZENN, a new kind of AI model, helping computers make sense of real-world information. Courtesy of Penn State/Jennifer M. McCann. Examples of energy compounds predicted by machine learning models trained by the Materials Project. Courtesy of Nature Materials.
RkJQdWJsaXNoZXIy MTYyMzk3NQ==