Results 71 to 80 of about 41,575 (296)

Neutrosophy-Driven Deep Learning for Predicting Student Performance [PDF]

Neutrosophic Sets and Systems
This paper proposes a hybrid architecture using several deep learning models in the neutrosophy environment for predicting student learning outcomes. The proposed framework proceeds on deep neural network models with the neutrosophy encoder/decoder.
N.T.K Son, N.T. Thong, N.H. Quynh
doaj   +1 more source

Recent Advances of Slip Sensors for Smart Robotics

Advanced Materials Technologies, EarlyView.
This review summarizes recent progress in robotic slip sensors across mechanical, electrical, thermal, optical, magnetic, and acoustic mechanisms, offering a comprehensive reference for the selection of slip sensors in robotic applications. In addition, current challenges and emerging trends are identified to advance the development of robust, adaptive,
Xingyu Zhang, Liao Wu, Shuai He, Zhao Sha, Yuwan Huang, Shuying Wu, Dewei Chu, Chun H. Wang, Shuhua Peng   +8 more
wiley   +1 more source

A Hybrid of Deep CNN and Bidirectional LSTM for Automatic Speech Recognition

Journal of Intelligent Systems, 2019
Deep neural networks (DNNs) have been playing a significant role in acoustic modeling. Convolutional neural networks (CNNs) are the advanced version of DNNs that achieve 4–12% relative gain in the word error rate (WER) over DNNs.
Passricha Vishal, Aggarwal Rajesh Kumar
doaj   +1 more source

Tunable Dynamics via Dual‐Ion Modulation for Event‐based Data Processing Using a Highly Uniform and Self‐Rectifying Memristor Array

Advanced Science, EarlyView.
Tunable dynamics of an interface‐type memristor array enabled by dual‐ion modulation through Ag nanoclusters are demonstrated. A 32 × 32 one‐resistor (1R) array exhibits 100% yield with high temporal/spatial uniformity (<3%) and a rectification ratio of 105.
Yoonho Cho, Dawon Kim, Jeonghong Lee, Dae‐won Kim, See‐On Park, Jongmin Bae, Taehwan Jang, Shinhyun Choi   +7 more
wiley   +1 more source

Mechanism‐Informed Machine Learning Enables Discovery of Oncolytic Peptides for Cancer Immunotherapy

Advanced Science, EarlyView.
MISPOP integrates ensemble learning with membrane‐active physicochemical priors to identify Dermaseptin‐S9, a natural oncolytic peptide that disrupts tumor membranes, triggers immunogenic cell death, and shows strong antitumor activity. The study illustrates a mechanism‐informed route from peptide sequence data to cancer immunotherapy leads.
Wen Zhang, Shengxin Lu, Guangyong Zheng, Shensuo Li, Hongyu Chen, Mei Hong, Xiangru Zhou, Ruotian Tang, Ye Wu, Weidong Zhang, Dong Lu, Xin Luan   +11 more
wiley   +1 more source

Edge-Fog-Cloud Distributed Architecture for Intelligent DDoS Detection and Mitigation

Cybernetics and Information Technologies
Cloud and distributed infrastructures face significant challenges from increasingly sophisticated Distributed Denial-of-Service (DDoS) attacks. Real-time efficiency is limited by the latency and scalability issues that affect traditional centralized ...
Sabrine Hedjaz, Abderrahmane Baadache, Fouzi Semchedine   +2 more
doaj   +1 more source

Surface Fermi Level Modulation of Photoanode by Optimized Conducting Nanoparticle Heterointerfaces for Enhanced Photoelectrochemical Water Splitting

Advanced Science, EarlyView.
Optimizing the heterointerface coverage of the conducting Ni2P nanoparticles on the surface of BVO photoanode manipulates the surface Fermi‐level modulation, which significantly enhances photoelectrochemical oxygen evolution reaction. Our multiscale simulations and experimental results reveal that an optimal Ni2P coverage of 9.4% pins the surface Fermi
Phuong Thi Pham, Seulgi Ji, Unbeom Baeck, Yuankai Li, Duong Nguyen Nguyen, Won Tae Hong, Yang You, John Moraros, Kyoungsuk Jin, Jun Young Lee, Chan‐Hwa Chung, Tae‐Hoon Kim, Jongwook Park, Heechae Choi, Jung Kyu Kim   +14 more
wiley   +1 more source

How Advanced Artificial Intelligence Technologies Shape Drug–Drug and Drug–Target Interaction Modeling

Advanced Science, EarlyView.
This review explores the convergence of artificial intelligence technologies in modeling drug–drug and drug–target interactions. By evaluating advanced feature engineering, architectural innovations, and learning paradigms reveals shared evolutionary trends and critical challenges, such as cold‐start settings and shortcut learning.
Xin Sun, Tong Wang
wiley   +1 more source

Training trajectory of DNN loss functions.

, 2022
Trajectory of training loss and test loss through training DNN surrogate for fitting objective function of optimal rotor profile problem. In the end of training, the test loss is significantly larger than the training loss, indicating the DNN training ...
Yaoyu Zhang (524584), Lulu Zhang (104611), Zhi-Qin John Xu (12915016)   +2 more
core   +2 more sources

Exploring non-zero position constraints: algorithm-hardware co-designed DNN sparse training method

Xibei Gongye Daxue Xuebao
On-device learning enables edge devices to continuously adapt to new data for AI applications. Leveraging sparsity to eliminate redundant computation and storage usage during training is a key approach to improving the learning efficiency of edge deep ...
WANG Miao, ZHANG Shengbing, ZHANG Meng
doaj   +1 more source