A Novel Pattern Recognition Method for Self-Powered TENG Sensor Embedded to the Robotic Hand by Azat Balapan, Rauan Yeralkhan, Alikhan Aryslanov, Gulnur Kalimuldina, Azamat Yeshmukhametov

“…These findings underscore the potential of integrating TENG sensors with advanced neural network architectures for autonomous dexterous manipulation in various industrial applications, paving the way for future advancements in robotic tactile sensing.…”
Get full text

Deep Learning in Music Generation: A Comprehensive Investigation of Models, Challenges and Future Directions by Kong Xiangchen

“…This review explores various and recent deep learning models, such as Long Short-Term Memory (LSTM) networks, Transformer-based models, Reinforcement Learning (RL), and Diffusion-based architectures, and how they are applied to music generation. …”
Get full text

Convolutional neural network prediction of the particle size distribution of soil from close-range images by Enrico Soranzo, Carlotta Guardiani, Wei Wu

“…We implemented the model in Python and explored various neural network architectures, image resolutions and data augmentation techniques to optimise performance. …”
Get full text

PBVit: A Patch-Based Vision Transformer for Enhanced Brain Tumor Detection by Pratikkumar Chauhan, Munindra Lunagaria, Deepak Kumar Verma, Krunal Vaghela, Ghanshyam G. Tejani, Sunil Kumar Sharma, Ahmad Raza Khan

“…The promising results demonstrate that PBVit can play a important role in facilitating early-stage diagnosis, reducing unnecessary biopsies, and ultimately enhancing patient care, while also showcasing the potential of transformer-based architectures in medical imaging.…”
Get full text

Biologically plausible gated recurrent neural networks for working memory and learning-to-learn. by Alexandra R van den Berg, Pieter R Roelfsema, Sander M Bohte

“…While meta-learning can be implemented in recurrent neural networks, these networks tend to be trained with architectures that are not easily interpretable or mappable to the brain and with learning rules that are biologically implausible. …”
Get full text

Extracting equations of motion from superconducting circuits by Christian Z. Pratt, Kyle J. Ray, James P. Crutchfield

“…This leads to efficient and practical circuit prototyping, as well as accessing scalable circuit architectures. The analytical efficiency is demonstrated by reproducing the potential energy landscape generated by a SQUID. …”
Get full text

Optimizing Deep Learning Acceleration on FPGA for Real-Time and Resource-Efficient Image Classification by Ahmad Mouri Zadeh Khaki, Ahyoung Choi

“…This work presents an optimized implementation of VGG16 and VGG19, two widely used CNN architectures, for classifying the CIFAR-10 dataset using transfer learning on field-programmable gate arrays (FPGAs). …”
Get full text

Tracking Vehicle Cruising in an Open Parking Lot Using Deep Learning and Kalman Filter by Vijay Paidi, Hasan Fleyeh, Johan Håkansson, Roger G. Nyberg

“…The vehicle is detected using Yolov3, modified Yolo, and custom Yolo deep learning architectures. The detected vehicles are tracked using Kalman filter and the trajectory of multiple vehicles is calculated on an image. …”
Get full text

Optimizing multi label student performance prediction with GNN-TINet: A contextual multidimensional deep learning framework. by Xiaoyi Zhang, Yakang Zhang, Angelina Lilac Chen, Manning Yu, Lihao Zhang

“…The GNN-TINet utilizes InceptionNet, transformer architectures, and graph neural networks (GNN) to improve precision in multi-label student performance forecasting. …”
Get full text

Multi-bandwidth reconstruction for photoacoustic tomography using cascade U-net by Zezheng Qin, Lingyu Ma, Zhigang Lei, Yiming Ma, Weiwei Fu, Mingjian Sun

“…Utilizing the two semi-ring array PAIs, we have devised a specialized deep learning model, comprising two serially connected U-net architectures, to autonomously generate multi-bandwidth full-view PAIs. …”
Get full text

A radix‐8 modulo 2n multiplier using area and power‐optimized hard multiple generator by Naveen Kr. Kabra, Zuber M. Patel

“…The synthesis of the proposed architectures is done using the Cadence tool at Generic Process design Kit‐45 nm technology. …”
Get full text

A High-Level Synthesis Scheduling and Binding Heuristic for FPGA Fault Tolerance by David Wilson, Aniruddha Shastri, Greg Stitt

“…In this paper, we present a more transparent high-level synthesis approach that uses scheduling and binding to provide attractive tradeoffs between area, performance, and redundancy, while focusing on FPGA implementation considerations, such as resource realization costs, to produce more efficient architectures. Compared to TMR applied to existing RTL, our approach shows resource savings up to 80% with average resource savings of 34% and an average clock degradation of 6%. …”
Get full text

Computing native network (CNN) technology and application driven by joint communication ＆ computing (JC＆C) by Jie SUN, Leiming MA, Aidong YANG, Xiaozhou YE, Peng LYU, Zhigang WANG, Ye OUYANG

“…The global telecom industry is carrying out research and exploration of potential 6G technical architectures and key techs, of which it has become a consensus in the industry to further realize the vision of 6G smart inclusion by better enabling all new 6G network technologies and services through joint communication ＆amp; computing (JC＆amp;C).Therefore, how to realize the deep integration and real-time collaboration of communication and computing to meet the needs of ultra-low latency and real-time response of new services in dynamic and complex network environments has become the challenge and key of JC＆amp;C.A computing native network (CNN) driven by JC＆amp;C was proposed, which realized the decoupling of communication and computing of base station or multiple base stations, computing pooling and intelligent JC＆amp;C scheduling through computing time division multiplexing and intelligent JC＆amp;C scheduling, thus realizing the base station to support communication and computation services simultaneously.The technology has been technically verified and piloted, and the experimental and pilot results demonstrate that CNN can effectively improve the computing utilization rate of wireless networks and realize the base station to support communication and computing services simultaneously, facilitating the evolution of 6G JC＆amp;C.…”
Get full text

Textural analysis and artificial intelligence as decision support tools in the diagnosis of multiple sclerosis – a systematic review by Filip Orzan, Ştefania D. Iancu, Laura Dioşan, Zoltán Bálint

“…The study emphasizes common models, including U-Net, Support Vector Machine, Random Forest, and K-Nearest Neighbors, alongside their evaluation metrics.ResultsThe analysis revealed a fragmented research landscape, with significant variation in model architectures and performance. Evaluation metrics such as Accuracy, Dice score, and Sensitivity are commonly employed, with some models demonstrating robustness across multi-center datasets. …”
Get full text

Multimodal Autism Spectrum Disorder Method Using GCN With Dual Transformers by Tianming Song, Zhe Ren, Jian Zhang, Yawei Qu, Yingying Cui, Zhengda Liang

“…This study proposes a novel diagnostic approach that integrates Graph convolutional networks (GCN) with dual transformer architectures, optimized through a Co-training strategy. …”
Get full text

Deep Learning Framework for Advanced De-Identification of Protected Health Information by Ahmad Aloqaily, Emad E. Abdallah, Rahaf Al-Zyoud, Esraa Abu Elsoud, Malak Al-Hassan, Alaa E. Abdallah

“…This rigorous approach to architectural design and hyperparameter tuning, often underexplored in the existing literature, significantly enhances the model’s capacity for accurate PHI tag detection while preserving the essential clinical context. …”
Get full text

An MLLM-Assisted Web Crawler Approach for Web Application Fuzzing by Wantong Yang, Enze Wang, Zhiwen Gui, Yuan Zhou, Baosheng Wang, Wei Xie

“…Web application fuzzing faces significant challenges in achieving comprehensive test interface (attack surface) coverage, primarily due to the complexity of user interactions and dynamic website architectures. While web crawlers can automatically access and extract critical website information—including form fields and request parameters—which are essential for generating effective fuzzing test cases, current crawler technologies exhibit three primary limitations: (i) insufficient capabilities in analyzing page relationships and determining page states; (ii) lack of functionality-aware exploration capabilities, resulting in generated inputs with poor contextual relevance; (iii) generation of unstructured operation sequences that fail to execute effectively due to their incompatibility with state-based testing logic. …”
Get full text

The Comparison of Using Andesite Waste Powder and Fly Ash in Concrete by Hakan Ceylan, Metin Davraz

“…Andesite stone is widely used in civil engineering and architectural applications in Turkey. The production, cutting and polishing of the andesite stone by mining methods causes a large amount of waste. …”
Get full text

Batch-in-Batch: a new adversarial training framework for initial perturbation and sample selection by Yinting Wu, Pai Peng, Bo Cai, Le Li

“…Through extensive experiments on three benchmark datasets with two network architectures in both single-step (Noise-Fast Gradient Sign Method, N-FGSM) and multi-step (Projected Gradient Descent, PGD) scenarios, models trained within the BB framework consistently demonstrate superior adversarial accuracy across various adversarial settings, notably achieving an improvement of more than 13% on the SVHN dataset with an attack radius of 8/255 compared to N-FGSM. …”
Get full text

Performance Analysis of Novel Lightweight Photovoltaic Curtain Wall Modules Under Different Climatic Conditions by Guofeng Tao, Qunzhi Zhu, Yonggang Chen, Xiaolong Shi, Jun Guan

“…Simulations were carried out to determine the power generation of faux architectural material PV curtain wall modules (FAM PVCWMs) for the best cavity distance per facade in various cities. …”
Get full text