Results 241 to 250 of about 142,362 (315)

Population-based variance-reduced evolution over stochastic landscapes. [PDF]

Sci Rep
Pei Z, He X, Pan Y, Peng B, Chen W, Zhou Y.   +5 more
europepmc   +1 more source

A Zeroth-Order Stochastic Gradient Descent Method for Communication-Efficient Federated Learning

Hodaka Nishi, Shiro Yano, Megumi Miyashita, Shunta Onishi, Yuta Goto, Toshiyuki Kondo   +5 more
openalex   +1 more source

In Situ Quantization with Memory‐Transistor Transfer Unit Based on Electrochemical Random‐Access Memory for Edge Applications

Advanced Science, EarlyView.
By combining ionic nonvolatile memories and transistors, this work proposes a compact synaptic unit to enable low‐precision neural network training. The design supports in situ weight quantization without extra programming and achieves accuracy comparable to ideal methods. This work obtains energy consumption advantage of 25.51× (ECRAM) and 4.84× (RRAM)
Zhen Yang, Yuxiang Yang, Baiqian Wang, Yaoyu Tao, Zelun Pan, Lei Cai, Teng Zhang, Longhao Yan, Xianbin Li, Yuchao Yang   +9 more
wiley   +1 more source

Enhanced Polarization-Maintaining Integrated Optical Beam Splitter Design via Stochastic Gradient Descent on Nanophotonic Waveguide Geometries

Freederia AI Researcher
openalex   +2 more sources

Accelerated Stochastic Conjugate Gradient for a class of convex optimization. [PDF]

PLoS One
He L, Du Y.
europepmc   +1 more source

Combining Electrochemical Reduction with Biosynthesis for Directed Conversion of CO2 into a Library of C3 Chemicals

Advanced Science, EarlyView.
In the H‐type electrolytic cell, carbon dioxide is reduced to acetic acid via electro‐microbial catalysis. The simply processed acetic acid is further converted through biological fermentation into high‐value‐added products, including acrylic acid, L‐lactic acid, and β‐alanine.
Kaixing Xiao, Shanquan Liang, Xujun Zhao, Zhiyao Peng, Ruoshi Luo, Jikai Zong, Ling Zhou, Yude Su, Dan Wang   +8 more
wiley   +1 more source

Dynamic differential privacy technique for deep learning models. [PDF]

Sci Rep
Elabd E.
europepmc   +1 more source

Empowering Carbon Fibers With Ti3C2Tx MXene: A Paradigm Shift Toward Integrated Structure‐Function Composites

Advanced Science, EarlyView.
This review comprehensively outlines how Ti3C2Tx MXene transforms carbon fiber from a structural component into a multifunctional platform. We systematically detail cutting‐edge modification strategies and showcase exceptional performance in EMI shielding, energy storage, smart sensing, and beyond.
Hongshuo Cao, Yue Xing, Jiangman Sun, Yanhong Tian, Yangyang Gao, Xuejun Zhang, Xiubing Liang   +6 more
wiley   +1 more source

The Price Equation Reveals a Universal Force-Metric-Bias Law of Algorithmic Learning and Natural Selection. [PDF]

Entropy (Basel)
Frank SA.
europepmc   +1 more source

Tri‐Layer Solid‐State Nanopore Arrays with Crosstalk Suppression for High‐Throughput, Femtomolar‐Level Biosensing

Advanced Science, EarlyView.
The critical pore spacing—approximately twenty times the pore radius—required to minimize electric field coupling is determined via finite element analysis. A multi‐layered Al2O3/Au/Si3N4 nanopore structure fabricated by helium ion beam lithography is proposed, enabling quantitative analysis of the target analyte while mitigating inter‐pore crosstalk ...
Silu Feng, Qinglong Luo, Siqi Ai, Suiwei Shen, Chengyong Wang, Zhishan Yuan   +5 more
wiley   +1 more source