Conference Logo ProgramMaster Logo
Conference Tools for 2027 TMS Annual Meeting & Exhibition
Login
Register as a New User
Help
Submit An Abstract
Propose A Symposium
Presenter/Author Tools
Organizer/Editor Tools

About this Abstract

Meeting 2027 TMS Annual Meeting & Exhibition
Symposium Hume-Rothery Symposium: Data-Driven Materials Discovery and Phase Stability
Presentation Title Calibrated Machine-Learning Uncertainty for Data-Driven Alloy Design
Author(s) Dongwon Shin
On-Site Speaker (Planned) Dongwon Shin
Abstract Scope Machine learning surrogates now predict composition-processing-property relationships accurately; however, they cannot state where their own recommendations should be trusted. We show that even a single, internally consistent experimental dataset, the most favorable case for trusting model uncertainty, yields ensemble intervals that are over-confident, and that locally-weighted conformal calibration restores empirically verified coverage at negligible cost. Applied to cast aluminum alloys, where composition and T5 heat treatment must be chosen together under process-window and cost constraints, calibration converts a cascaded surrogate into a decision tool. Calibrated intervals separate validated recommendations from exploration targets, and reveal that upper-confidence-bound optimization drifts into the region where its own guarantee fails, justifying a mean-based Pareto formulation with interval reporting. We then co-optimize hardness, electrical conductivity, and processing cost, mapping calibrated Pareto fronts across three heat-treatment scenarios, which show that process-window width governs the reachable hardness-conductivity compromises.
Proceedings Inclusion? Planned:

OTHER PAPERS PLANNED FOR THIS SYMPOSIUM

A Computational Thermodynamics Framework with Intrinsic Chemical Short-Range Order
Accelerating Oxygen Ion Conductor Discovery Through First-Principles Design and Autonomous Materials Optimization
AI in the Wild: Autonomous laboratories for materials synthesis
Alloy theory applied to the anodes of all-solid-state batteries
Alloying effects on deformation modes in wrought magnesium: An integrated computational and experimental investigation
Atomic-Scale Simulations of Metal/Molten-Salt Interfaces by Machine-Learning Interatomic Potentials
Beyond Order: A Disorder-Aware Workflow toward OQMD 2.0
Beyond Randomness: Designing Short-Range Order for Materials Performance
Bonding-Driven Discovery and Design of Thermoelectric Materials
Calibrated Machine-Learning Uncertainty for Data-Driven Alloy Design
Chemical Intuition as a Guide to Data-Driven Materials Discovery
Computational Design of Battery Guided by Degradation Mechanism
Computational discovery of materials for energy storage
Data- and Simulation-Driven Design of Structural and Functional Materials: Successes and Challenges
Data-Driven Discovery of Novel High-Performance Photovoltaics in an Experimentally Known Family of Quaternary Chalcogenides
Data-driven Prediction of Solid-State Synthesis
Elemental Features for Data-Driven Materials Properties Prediction
Elucidating the role of solute-solute interactions on diffusion in Ni-based substitutional solid solutions
Exploring the electronic structure of all known inorganics - from transport to superconductivity to topology.
First-principles modeling of disorder in cathodes and beyond
From Computational Design to Accelerated Discovery of High-Entropy Alloys
Functional Synthesizability: Tuning the Extreme-Properties of High-Entropy Ceramics
High-Throughput Anharmonic Phonons for Data-Driven Materials Discovery, Phase Stability, and Thermal Transport
How Computational Modeling Shapes Materials at Apple
Hume-Rothery Strikes Again: Intrinsic Correlations Permit Tailored Materials With Exceptional Properties
Integrating Automated Computation with Experiment for Accelerated Materials Discovery
Irreversible Thermodynamic Basis for the Phase-field Method of Ordered Stoichiometric Compounds
Lithium Extraction with Ion Exchange at Lilac Solutions
Mapping Lithium-Ion Transport in Sulfide Argyrodites: From Defect Chemistry to Composite Interfaces
Phase Diagrams On-Demand
Relationship Between Configurational and Vibrational Entropies of Mixing and Their Effects on Phase Diagrams
Standing on Chris's Shoulders: From High-Throughput DFT to AI for Materials Discovery
Symmetry Broken DFT Corrects the Stability and Mott Band Gap Errors Without Adding Strong Correlations
The Generalized Aliasing Decomposition, A New Paradigm for Modeling
The Possibility of New Complex Magnet Material
The Spatial Arrangement of Nanoparticles on Substrates and the Role of Phase Transformations
Theoretical Insights Into Hydrogenation and Proton Transport in ABO3 Perovskite
Thermodynamic and Kinetic Stability of M–N–C Active Sites for O₂ Reduction
Towards Ai-enabled High Throughput Characterization and Materials Discovery
Zentropy: A Thermodynamic Framework Bridging Phase Stability, Data-Driven Modeling, and Artificial Intelligence

Questions about ProgramMaster? Contact programming@programmaster.org | TMS Privacy Policy | Accessibility Statement