A new topology-based method predicts atomic charges in metal-organic frameworks from bond connectivity alone, making large-scale computational screening practical.

Abstract: Backpropagation neural networks are commonly utilized to solve complicated issues in various disciplines. However, optimizing their settings remains a significant task. Traditional ...

Abstract: Decentralized learning is a distributed learning approach, wherein clients collaborate to learn a global model. Many optimization problems that appear in signal processing and machine ...

This project implements a 2D pore-throat network extraction algorithm for porous media images. It uses a modified watershed segmentation approach based on Distance Transform and H-maxima markers to ...

Official implementation of GradES - a gradient-based selective training method that dynamically freezes converged modules during fine-tuning to achieve 40-50% computational savings without sacrificing ...