The keywords assigned to papers by authors in the field of affective computing were analyzed for frequency and co-occurrence, and the core keywords among them were clustered to get five…
Tag: Computer Sciences
Automated machine learning from the perspective of bilevel optimization
(a) Illustration of key issues of ML task. (b) Formulation of AutoML paradigm from the perspective of bilevel optimization. Credit: Science China Press Recently, professors Risheng Liu from Dalian University…
New system combines human, artificial intelligence to improve experimentation
The workflow of the human-AI collaborator system developed at ORNL to improve experimentation. Credit: Arpan Biswas and Rama Vasudevan/ORNL, U.S. Dept. of Energy Though artificial intelligence decreases human error in…
A theoretical model for reliability assessment of machine learning systems
Credit: Pixabay/CC0 Public Domain For a machine learning system comprising multiple machine learning models and input data, researchers at University of Tsukuba developed a theoretical model for evaluating the effect…
Researchers harness large language models to accelerate materials discovery
A scanning electron microscope image of tin crystals, stimulated by electricity and growing on a copper surface. A new method developed by Princeton researchers could speed up the process of…
What if American nuclear power plants could be less expensive
PINNs may support better predictions by fusing neural networks with differential equations to solve complex scientific or engineering problems. Credit: Argonne National Laboratory American nuclear power plants generate more than…
Researchers demonstrate scalability of graph neural networks on world’s most powerful computing systems
Conversion of an atomic structure into a graph, where atoms are treating as nodes and interatomic bonds as edges. Credit: Massimiliano “Max” Lupo Pasini/ORNL, U.S. Dept. of Energy Solving today’s…
The solution space of the spherical negative perceptron model is star-shaped, researchers find
Sketch of the negative perceptron’s solution landscape satisfying the required constraints (blue and white) or with nonzero energy (orange). Credit: Annesi et al. Recent numerical studies investigating neural networks have…
