Results 111 to 120 of about 124,232 (320)

Big Data and AI‐Powered Modeling: A Pathway to Sustainable Precision Animal Nutrition

Advanced Science, EarlyView.
This review summarizes the current landscape of big data and AI‐powered modeling in animal nutrition, covering techniques including intelligent data acquisition, data augmentation, explainable machine learning, heuristic algorithms, and life cycle assessment‐based sustainability evaluation.
Shuai Zhang, Changhua Lai, Jinbiao Zhao, Junjun Wang   +3 more
wiley   +1 more source

STEMDiff: A Wavelet‐Enhanced Diffusion Model for Physics‐Informed STEM Image Generation

Advanced Science, EarlyView.
STEMDiff, which is a conditional diffusion model that generates realistic STEM images from crystal structure‐derived binary labels, is proposed, overcoming high‐frequency bias via a Discrete Wavelet Transform‐based architecture. It produces experimentally indistinguishable images, enabling annotation‐free training of networks for precise atomic ...
Yihui Bao, Xinyi Lu, Yanyan Xia, Zhencheng Ye, Houyang Chen   +4 more
wiley   +1 more source

A progressive growing of conditional generative adversarial networks model

Dianxin kexue, 2023
Progressive growing of generative adversarial networks (PGGAN) is an adversarial network model that can generate high-resolution images.However, when the categories of samples are unbalanced, or the categories of samples are too similar or too dissimilar,
Hui MA, Ruiqin WANG, Shuai YANG
doaj   +2 more sources

Multiscale Cell–Cell Interactive Spatial Transcriptomics Analysis

Advanced Science, EarlyView.
In this study, we present the MultiScale Cell‐Cell Interactive Spatial Transcriptomics Analysis method, which unites the strengths of spatially resolved deep learning techniques with a topological representation of multi‐scale cell‐cell similarity relations.
Sean Cottrell, Guo‐Wei Wei
wiley   +1 more source

A Novel GAN-based Fault Diagnosis Approach for Imbalanced Industrial Time Series

, 2019
This paper proposes a novel fault diagnosis approach based on generative adversarial networks (GAN) for imbalanced industrial time series where normal samples are much larger than failure cases.
Cheng, Cheng, Jiang, Wenqian, Ma, Guijun, Yuan, Ye, Zhou, Beitong   +4 more
core  

High‐Asymmetry Metasurface: A New Solution for Terahertz Resonance via Active Learning‐Augmented Diffusion Model

Advanced Science, EarlyView.
A prior knowledge‐guided diffusion model augmented by physics‐constrained active learning is developed to design high‐asymmetry terahertz metamaterials. Trained on only a small set of classical structures, the model efficiently generates new high‐metrics designs.
Qiqi Dai, Yinpeng Wang, Cheng Xu, Dongxiao Li, Prakash Pitchappa, Thomas Caiwei Tan, Ranjan Singh, Chengkuo Lee   +7 more
wiley   +1 more source

InfoGAN: Interpretable Representation Learning by Information Maximizing Generative Adversarial Nets

, 2016
This paper describes InfoGAN, an information-theoretic extension to the Generative Adversarial Network that is able to learn disentangled representations in a completely unsupervised manner. InfoGAN is a generative adversarial network that also maximizes
Abbeel, Pieter, Chen, Xi, Duan, Yan, Houthooft, Rein, Schulman, John, Sutskever, Ilya   +5 more
core  

SpaBatch: Deep Learning‐Based Cross‐Slice Integration and 3D Spatial Domain Identification in Spatial Transcriptomics

Advanced Science, EarlyView.
SpaBatch is an end‐to‐end multi‐slice spatial transcriptomics data integration framework. It simultaneously performs embedding learning, spatial feature denoising and reconstruction, batch effect correction, and spatial domain optimization, effectively correcting batch effects and achieving accurate 3D spatial domain identification.
Jinyun Niu, Donghai Fang, Jinyu Chen, Yi Xiong, Juan Liu, Wenwen Min   +5 more
wiley   +1 more source

Generative Adversarial Networks Unlearning

IEEE Transactions on Dependable and Secure Computing
As machine learning continues to develop, and data misuse scandals become more prevalent, individuals are becoming increasingly concerned about their personal information and are advocating for the right to remove their data. Machine unlearning has emerged as a solution to erase training data from trained machine learning models. Despite its success in
Hui Sun, Tianqing Zhu, Wenhan Chang, Wanlei Zhou   +3 more
openaire   +2 more sources

Re‐Purposing a Modular Origami Manipulator Into an Adaptive Physical Computer for Machine Learning and Robotic Perception

Advanced Science, EarlyView.
Reservoir computing in the physical domain promises an alternative, more energy‐efficient approach to machine learning. However, the correlation between mechanical design and computing power remains elusive. This study sheds light on this gap by repurposing a meta‐stable modular origami manipulator into an adaptive computing kernel that can perform ...
Jun Wang, Suyi Li
wiley   +1 more source