Search Results for author:
Found 2 papers, 1 papers with code
EDiT: A Local-SGD-Based Efficient Distributed Training Method for Large Language Models
Local SGD methods have been proposed to address these issues, but their effectiveness remains limited to small-scale training due to additional memory overhead and lack of concerns on efficiency and stability.
Machine Learning Regression based Single Event Transient Modeling Method for Circuit-Level Simulation
A multilayer feedfordward neural network is used to build the SET pulse current model by learning the data from TCAD simulation.
Cannot find the paper you are looking for? You can
Submit a new open access paper.
