Tag: Machine Learning
-
Key structures to metallic glass stability revealed with machine learning
Using the second-nearest neighboring atoms to predict metallic glass stability can help researchers more accurately model the disordered solid with strong, elastic properties.
-
Providing the Artemis mission with solar radiation forecasts
Machine-learning and physics-based models developed at U-M will warn NASA when solar particle radiation could become hazardous up to 24 hours in advance.
-
How to improve AI energy efficiency with open-source tools: Q&A with Mosharaf Chowdhury
Any company could use our tools to measure and optimize their AI models and reduce AI energy use.
-
Letting atomic simulations learn from phase diagrams
Ten times more efficient than previous methods, a new machine learning method builds a two-way connection between atomic simulation and experimental data.
-
AI tool predicts battery cycle life with just a few days’ data
A ‘learner,’ ‘interpreter’ and ‘oracle’ work together with minimal experiments to draw parallels between historical data and new battery designs.
-
AI supports home-based balance training
New machine learning model draws data from wearable sensors to predict how a physical therapist would assess balance training performance.
-
Interpretable machine learning to accelerate nanocatalyst discovery
A fast and accurate surrogate model screens over 10,000 possible metal-oxide supports for a platinum nanocatalyst to prevent sintering under high temperatures.
-
AI for studying turbulence: A fresh look at an unsolved physics problem
Explainable AI helps find key drivers of turbulence, offering new insights that could improve flight safety and industrial efficiency.
-
Between rain and snow, machine learning finds 9 precipitation types
Leveraging 1.5M minutes of precipitation data and a nonlinear method to handle complex relationships between variables, the team created a new classification system
-
Quantum chemistry: Making key simulation approach more accurate
Density functional theory is limited by a mystery at its heart: the universal exchange-correlation functional. U-M researchers are trying to uncover it.
-
AI system discovers visual categories while adapting to new contexts
Open ad-hoc categorization approach combines language guidance with visual clustering to learn contextualized features for flexible image interpretation.
-
Improving AI models: Automated tool detects silent errors in deep learning training
TrainCheck uses training invariants to find the root cause of hard-to-detect errors before they cause downstream problems, saving time and resources.