Results 71 to 80 of about 384,953 (220)

Iterative regularization for low complexity regularizers

Numerische Mathematik
Iterative regularization exploits the implicit bias of an optimization algorithm to regularize ill-posed problems. Constructing algorithms with such built-in regularization mechanisms is a classic challenge in inverse problems but also in modern machine learning, where it provides both a new perspective on algorithms analysis, and significant speed-ups
Molinari C., Massias M., Rosasco L., Villa S.   +3 more
openaire   +2 more sources

Machine Learning‐Driven Multi‐Objective Optimization of Microchannel Reactors for CO₂ Conversion

Advanced Sustainable Systems, EarlyView.
This study introduces a novel method that combines CFD, RSM, and ML to improve a microreactor's performance utilizing the Sabatier reaction. A range of ML models is assessed, and the best one is selected to predict optimal reactor conditions. ML shows the ability to predict performance in just milliseconds, leading to a decrease in computational time ...
Sandeep Kumar, Parmod Kumar, Kuljeet Singh Grewal   +2 more
wiley   +1 more source

Accurate and Efficient Behavioral Modeling of GaN HEMTs Using An Optimized Light Gradient Boosting Machine

Advanced Theory and Simulations, EarlyView.
Machine Learning (ML) and optimization have permeated almost every aspect of engineering applications. Recent years have seen great traction toward ML‐based GaN HEMT modelling. However, ML‐based GaN HEMT models are mostly developed using variants of Artificial Neural Network (ANN).
Saddam Husain, Mohammad Hashmi, Fadhel M. Ghannouchi   +2 more
wiley   +1 more source

Predicting Fiber Length Characteristics of Recycled Cotton and Cellulose Fiber Blends Using Machine Learning Models

Advanced Theory and Simulations, EarlyView.
This study explores machine learning‐driven prediction of fiber length characteristics in sustainable yarn blends made from recycled cotton and Lyocell. By analyzing empirical data through models like Random Forest and Gradient Boosting, and interpreting results with SHAP, key fiber length features from the Staple Diagram and Fibrogram are identified ...
Tuser Tirtha Biswas, Michael Will, Nawar Kadi   +2 more
wiley   +1 more source

Integrative Multi‐Omics and Routine Blood Analysis Using Deep Learning: Cost‐Effective Early Prediction of Chronic Disease Risks

Advanced Science, EarlyView.
Omicsformer, a deep learning model, integrates multi‐omics and routine blood data to accurately predict risks for nine chronic diseases, including cancer and cardiovascular conditions. Validated using large scale clinical data, it reveals early risk trajectories, advancing personalized medicine and offering a cost‐effective, community‐based solution ...
Zhibin Dong, Pei Li, Yi Jiang, Zhihan Wang, Shihui Fu, Hebin Che, Meng Liu, Xiaojing Zhao, Chunlei Liu, Chenghui Zhao, Qin Zhong, Chongyou Rao, Siwei Wang, Suyuan Liu, Dayu Hu, Dongjin Wang, Juntao Gao, Kai Guo, Xinwang Liu, En Zhu, Kunlun He   +20 more
wiley   +1 more source

Exploring the Latent Information in Spatial Transcriptomics Data via Multi‐View Graph Convolutional Network Based on Implicit Contrastive Learning

Advanced Science, EarlyView.
STMIGCL is an implicit contrastive learning‐based multi‐view graph convolutional network framework designed for downstream tasks such as spatial domain recognition, trajectory inference, and spatially variable gene identification. By combining multi‐view learning with contrastive learning and employing contrastive learning methods that enhance contrast
Sheng Ren, Xingyu Liao, Farong Liu, Jie Li, Xin Gao, Bin Yu   +5 more
wiley   +1 more source

Regularity at the Boundary and Tangential Regularity

, 2011
For a pseudoconvex domain in complex space, we prove the equivalence of the local hypoellipticity of the system (di-bar, di-bar*) with the system (di-bar_b,di-bar*_b) induced in the boundary. This develops a result of ours which used the theory of the "harmonic" extension by Kohn.
Khanh, Tran Vu, Zampieri, Giuseppe
openaire   +2 more sources

High‐Fidelity Computational Microscopy via Feature‐Domain Phase Retrieval

Advanced Science, EarlyView.
An innovative phase retrieval framework, termed FD‐PR, is uniquely established in the image's feature domain through the feature‐extracted, physical‐driven regression with interfaces for combining physics and image processing constraints. FD‐PR takes advantage of invariance components of an image against presences of model mismatch and uncertainty ...
Shuhe Zhang, An Pan, Hongbo Sun, Yidong Tan, Liangcai Cao   +4 more
wiley   +1 more source

Regular factors in regular multipartite graphs

Discrete Mathematics, 2000
AbstractWe present sufficient conditions for a regular multipartite graph to have a regular factor and show that these are best possible.
openaire   +3 more sources

Regular Partitions of Regular Graphs [PDF]

Canadian Mathematical Bulletin, 1978
In the study of the combinatorial structure of edge-graphs of convex polytopes one may ask whether a given graph possesses a partition consisting of certain kinds of subgraphs.In this paper we describe some special partitions of 3-valent and 4-valent graphs.
openaire   +2 more sources