As artificial intelligence reshapes how materials are discovered, designed, and qualified, the ability to learn from data has never been more central to materials science and engineering. This conference brings together leading experts, researchers, and industry professionals to discuss the latest advancements, challenges, and opportunities in the following topics:
Focuses on using image-based techniques to reconstruct and analyze microstructures.
Covers methods for image reconstruction, enhancement, registration, denoising, super resolution, stitching and quality improvement. This can also include multimodal or correlative imaging approaches combining information from different microscopy techniques.
Covers methods like segmentation, classification, defect detection, phase identification and oval quantitative characterization of microstructures, including generative methods for generating synthetic microstructures.
Focuses on AI methods for analyzing signal-based microstructure characterization and process data such as XRD, spectroscopy, SAM / acoustic signals, thermal signals or other sensor data.
Covers machine-learning methods for extracting information from signals and possibly combining multiple techniques to analyze microstructures or materials processing.
Involves using AI to predict material properties and structures based on their microstructural characteristics, composition, processing conditions or simulation data.
Discusses the application of machine learning to support, accelerate or complement materials simulations across different length and time scales. This includes atomistic, quantum-mechanical, continuum, finite-element and phase-field simulations, as well as machine-learning potentials, surrogate models, reduced-order models and parameter identification.
Focuses on using machine learning to describe and predict relationships between material composition, processing, microstructure and resulting properties. Applications include property prediction, microstructure evolution, degradation, performance, and others.
Discusses machine-learning approaches that incorporate physical laws, constraints or simulations, improving prediction accuracy and physical consistency for scientific problems. This includes physics-informed ML as well as broader hybrid methods.
Explores how AI can accelerate the discovery of new materials with desired properties.
Examines adaptive machine-learning techniques where the algorithm actively queries information to learn more effectively. This includes active learning, Bayesian optimization and high-throughput experimental or computational screening.
Covers approaches that identify suitable materials, compositions, structures or processing conditions based on desired target properties. This can include inverse models, optimization approaches and generative AI for proposing new material candidates.
Focuses on the introduction and potential impact of novel data science technologies in material science and engineering.
Involves the use of AI to process and analyze human language and written text within the context of material science. This also includes foundation models and multimodal models that combine text with images, spectra or other scientific data.
Addresses the use of agents for materials science applications including device setup and control, automating data analysis and developments towards self-driving labs. This also includes autonomous experiment planning and closed-loop scientific workflows.
Focuses on the use of AI to optimize processing conditions in order to achieve desired material properties, microstructures or manufacturing outcomes.
Covers machine-learning methods for optimizing process parameters, monitoring process conditions and controlling manufacturing or materials-processing. Applications include predictive quality, anomaly detection and adaptive process control and related topics.
Covers the combination of simulations, sensor data and machine learning to create digital representations of materials. These methods can support real-time prediction, process optimization and adaptive manufacturing.
Focuses on the data foundations and methodological requirements needed for reliable and reusable AI in materials science.
Includes materials databases, ontologies, metadata, descriptors, learned representations and other approaches for working efficiently with materials data.
Covers uncertainty quantification, interpretability, robustness, benchmarking and validation of AI models. The focus is on understanding when AI predictions can be trusted and how models generalize to new materials, experiments or processing conditions.