Results 91 to 100 of about 226,838 (267)

Effect of terahertz irradiation on DNA damage repair in living cells

FEBS Open Bio, EarlyView.
We investigated the effect of terahertz (THz) wave irradiation on DNA double‐strand break (DSB) repair in living cells. We found that THz irradiation enhanced DSB repair at specific frequencies, whereas heat treatment inhibited it, indicating that this effect is nonthermal and frequency‐specific.
Yuya Ueno, Shota Yamazaki, Hiromichi Hoshina, Masahiko Harata   +3 more
wiley   +1 more source

A New Lossless Compression Paradigm for Federated Learning: A Quantile-Based Framework for Bandwidth Efficiency Without Accuracy Degradation

Information
Federated Learning (FL) is a machine learning technique that preserves data privacy and security by training models directly on decentralized edge network devices.
Marwa Abdellah, Aya Hesham, Ahmad Salah, Gamal M. Behery   +3 more
doaj   +1 more source

On the role of compressibility in poroviscoelastic models

Mathematical Biosciences and Engineering, 2019
In this article we conduct an analytical study of a poroviscoelastic mixture model stemming from the classical Biot's consolidation model for poroelastic media, comprising a fluid component and a solid component, coupled with a viscoelastic stress-strain relationship for the total stress tensor.
Bociu L., Guidoboni G., Sacco R., Verri M.   +3 more
openaire   +5 more sources

A new flow chip in combination with multiphoton microscopy as a protocol for longitudinal 3D imaging of tissue calcification under shear stress

FEBS Open Bio, EarlyView.
Miniaturized flow chip platform enabling continuous perfusion and longitudinal multiphoton 3D imaging of vascular smooth muscle cell constructs under physiological flow. Brightfield imaging guides region selection, while CellTracker Green and mRuby‐labeled fetuin‐A visualize cells and mineral deposition, respectively. Magnesium supplementation markedly
Vytautas Kučikas, Aaron D. Morgan, Ignas Čiurlionis, Christian Böhm, Stefan Jockenhövel, Willi Jahnen‐Dechent, Marc A. M. J. van Zandvoort   +6 more
wiley   +1 more source

Scaling Laws in the Tiny Regime: How Small Models Change Their Mistakes

Machine Learning and Knowledge Extraction
Neural scaling laws describe how model performance improves as a power law with size, but existing work has focused almost entirely on models above 100 M parameters.
Mohammed Alnemari, Rizwan Qureshi, Nader Bagherzadeh   +2 more
doaj   +1 more source

A Gradually Distilled CNN for SAR Target Recognition

IEEE Access, 2019
Convolutional neural networks (CNNs) have been widely used in synthetic aperture radar (SAR) target recognition. Traditional CNNs suffer from expensive computation and high memory consumption, impeding their deployment in real-time recognition systems of
Rui Min, Hai Lan, Zongjie Cao, Zongyong Cui   +3 more
doaj   +1 more source

HMC: Hybrid model compression method based on layer sensitivity grouping. [PDF]

PLoS One, 2023
Yang G, Yu S, Yang H, Nie Z, Wang J.
europepmc   +1 more source

A Survey on Model Compression for Large Language Models

Transactions of the Association for Computational Linguistics
Abstract Large Language Models (LLMs) have transformed natural language processing tasks successfully. Yet, their large size and high computational needs pose challenges for practical use, especially in resource-limited settings. Model compression has emerged as a key research area to address these challenges.
Xunyu Zhu, Jian Li 0040, Yong Liu 0018, Can Ma, Weiping Wang 0005   +4 more
openaire   +2 more sources

Evolution‐guided yeast complementation reveals functional differences in human PSPH variants

FEBS Open Bio, EarlyView.
Ancient genomes can help guide which human genetic variants are tested experimentally. This study applies that idea to PSPH, a gene involved in serine biosynthesis, and uses high‐throughput yeast complementation to compare variant function. The findings reveal measurable differences among selected alleles and illustrate the value of evolution‐guided ...
Mauricio Campa‐Álvarez, Diana Ascencio, Miguel Vallebueno‐Estrada, Eduardo González‐Orozco, Christian Eduardo Martínez‐Guerrero, Rafael Montiel, Alexander DeLuna   +6 more
wiley   +1 more source

Honest Teacher Knowledge Distillation: A Framework for Uncertainty-Aware Model Compression Using Gaussian Processes

IEEE Access
We introduce Honest Teacher Knowledge Distillation (HTKD), an uncertainty-aware distillation framework that transfers both knowledge and confidence from a teacher to a compact student by integrating Gaussian Process (GP)–based predictive ...
Arjay R. Alba, Jocelyn F. Villaverde
doaj   +1 more source