News
Chemist proposes shared “model proteins” to improve reproducibility in protein science
3+ hour, 47+ min ago (448+ words) image:This graphic outlines a proposed model protein system that pairs widely used reference proteins with shared benchmarks and reporting standards to support reproducible and comparable protein research. view more NEW LONDON, Conn. " Protein scientists could improve reproducibility and coordination…...
AI-based prediction of train-induced environmental vibration with limited measurements
5+ hour, 47+ min ago (380+ words) This study presents a transfer learning'based method for predicting train-induced environmental vibration. The method applies data fusion to combine physics-based numerical simulations and limited measurement data within a neural network. It reduces the heavy reliance of conventional machine learning'based models…...
RetinaDA: a diverse dataset for domain adaptation in retinal vessel segmentation
1+ day, 2+ hour ago (32+ words) image:Illustration of the RetinaDA dataset view more Illustration of the RetinaDA dataset Credit: HIGHER EDUCATON PRESS Frontiers of Computer Science RetinaDA: a diverse dataset for domain adaptation in retinal vessel segmentation...
FedCare: towards interactive diagnosis of federated learning systems
1+ day, 2+ hour ago (195+ words) image:Visual interface of FedCare view more Visual interface of FedCare Credit: HIGHER EDUCATON PRESS To solve the problems, a research team led by Tianye ZHANG published their new researchon 15 July 2025 in Frontiers of Computer Scienceco-published by Higher Education Press…...
Boosting recommendation performance by user intention aware visual feature pre-training
1+ day, 2+ hour ago (637+ words) Boosting recommendation performance by user intention aware visual feature pre-training'EurekAlert! User intention reconstruction module Credit: HIGHER EDUCATON PRESS With the advancement of multimedia internet, the impact of visual characteristics on the decision of users to click or not within the…...
Robust semi-supervised learning in open environments
1+ day, 2+ hour ago (200+ words) Robust semi-supervised learning in open environments'EurekAlert! Credit: HIGHER EDUCATON PRESS Semi-supervised learning (SSL) aims to improve performance by exploiting unlabeled data when labels are scarce. Conventional SSL studies typically assume close environments where important factors (e.g., label, feature, distribution) between labeled…...
Multi-constraint reinforcement learning in complex robot environments
1+ day, 2+ hour ago (234+ words) Multi-constraint reinforcement learning in complex robot environments'EurekAlert! Experimental results of FPMCO Credit: HIGHER EDUCATON PRESS Researchers have developed a novel multi-constraint optimization method that significantly improves the efficiency of reinforcement learning in complex environments. This new algorithm, called First-Order Projection-based…...
Physics of foam strangely resembles AI training
1+ day, 4+ hour ago (509+ words) Penn Engineers discover that bubble movements in foam mathematically mirror deep learning in AI systems image:A microscopic close-up of the bubbles in foam, whose movements mathematically mirror the process of deep learning, used to train modern AI systems. view…...
AI-accelerated gas sensing: a first-principles–guided machine learning platform for black phosphorus sensors
3+ day, 2+ hour ago (155+ words) Credit: Journal of Advanced Ceramics, Tsinghua University Press The team published their work in Journal of Advanced Ceramics on January 06, 2026. The corresponding author Guiwu Liu stated that: "while the current model is limited to black phosphorus, it is capable of…...
EduStudio: Towards a unified library for student cognitive modeling
1+ week, 4+ min ago (188+ words) Credit: HIGHER EDUCATON PRESS Recent efforts have been dedicated to the development of open-source code libraries for student cognitive modeling. However, existing libraries often focus on a particular category and overlook the relationships between them. Additionally, these libraries lack sufficient…...