Article
Computer Science, Artificial Intelligence
Yanli Yang, Yichuan He
Summary: An intelligent fault identification algorithm based on ensemble deep neural networks and correlation coefficients is proposed in this study, which can identify rotating machinery fault signals from multiple angles and make a comprehensive judgment by combining correlation coefficients, thus improving the reliability of fault identification.
Article
Acoustics
Luoyi Feng, Ming Zan, Linsen Huang, Zhongming Xu
Summary: Deep learning, based on deep neural networks, is widely used in various application fields and shows great promise in sound source identification. The grid-free method, a high-accuracy deep learning approach, requires predefining the number of sources. To overcome this limitation, a double-step grid-free method is proposed to identify an unknown number of sound sources.
Article
Computer Science, Interdisciplinary Applications
David Lehky, Martin Lipowczan, Hana Simonova, Zbynek Kersner
Summary: This paper presents a method for identifying parameters of fine-grained brittle matrix composites using artificial neural networks. An ensemble of artificial neural networks is created to cover the wide range of composite mixtures and achieve parameter values with sufficient precision. The system is also easily expandable to test composites with properties outside the current range.
COMPUTERS AND CONCRETE
(2021)
Article
Plant Sciences
Le Yang, Xiaoyun Yu, Shaoping Zhang, Huanhuan Zhang, Shuang Xu, Huibin Long, Yingwen Zhu
Summary: Rice leaf diseases significantly affect rice yields. This study proposes a stacking-based integrated learning model for efficient and accurate identification of rice leaf diseases. The model achieved a recognition rate of 99.69% on a rice dataset, outperforming single models and providing a better method for plant disease identification.
FRONTIERS IN PLANT SCIENCE
(2023)
Article
Biochemical Research Methods
Jiawei Li, Yuqian Pu, Jijun Tang, Quan Zou, Fei Guo
Summary: Quantifying DNA properties is a challenging task in human genomics, and understanding non-coding DNA functions is crucial for biological research. A hybrid deep neural network method, DeepATT, is proposed to identify regulatory functions on millions of DNA sequences, outperforming existing tools and reducing parameters while maintaining accuracy. The model captures regulatory motifs, grammar, and selects features efficiently, providing insights into DNA function correlations.
BRIEFINGS IN BIOINFORMATICS
(2021)
Article
Chemistry, Multidisciplinary
Mohammed Gamal Ragab, Said Jadid Abdulkadir, Norshakirah Aziz, Hitham Alhussian, Abubakar Bala, Alawi Alqushaibi
Summary: This paper introduces an end-to-end method for environmental sound classification based on deep learning, which directly learns features representation from the audio signal by dividing the signal into frames using a sliding window. Bayesian optimization is used for hyperparameter selection and model evaluation, achieving a high classification accuracy on the UrbanSound8K dataset.
APPLIED SCIENCES-BASEL
(2021)
Review
Engineering, Aerospace
Xi Chen, Lei Dong, Hong-Chang Li, Xin-Peng Yao, Peng Wang, Shuang Yao
Summary: Defects and errors in code pose a potential risk to software operation and require a proper code review process, especially for safety-critical software. The traditional manual review method is no longer sufficient due to the increasing size and variety of code. Deep Reviewer is a flexible framework that automatically detects code defects and correlates review comments. It achieves high precision and F1 scores and outperforms other methods in multi-classification tasks.
Article
Computer Science, Artificial Intelligence
Yongquan Yang, Haijun Lv, Ning Chen, Yang Wu, Jiayi Zheng, Zhongxi Zheng
Summary: Ensembles of deep CNNs play a crucial role in ensemble learning for artificial intelligence applications, but the increasing complexity of deep CNN architectures and large data dimensionality have made their usage costly. A new approach is proposed to find multiple models converging to local minima in the subparameter space of deep CNNs, which can improve generalization while being more affordable during training and testing stages.
PATTERN RECOGNITION
(2021)
Article
Chemistry, Multidisciplinary
Jin-A Lee, Keun-Chang Kwak
Summary: This paper proposes a method for heart sound classification using wavelet analysis techniques and an ensemble of deep learning models. The experimental results show that the proposed method performs well on two datasets compared to previous methods.
APPLIED SCIENCES-BASEL
(2023)
Article
Ecology
Thu Huong Truong, Huu Du Nguyen, Thi Quynh Anh Mai, Hoang Long Nguyen, Tran Nhat Minh Dang, Thi-Thu-Hong Phan
Summary: Bee buzzing carries important information about colony behavior and applying machine learning and deep learning techniques for monitoring beehives based on sound has gained interest worldwide.
ECOLOGICAL INFORMATICS
(2023)
Review
Computer Science, Information Systems
Irshad Ahmad Thukroo, Rumaan Bashir, Kaiser J. Giri
Summary: Information Technology has significantly impacted the field of Natural Language Processing, particularly in the area of spoken language identification. This paper investigates existing models for spoken language identification implemented using various deep learning approaches and highlights their features and challenges. A comprehensive comparative study of deep learning techniques has been conducted to analyze the efficiency of spoken language models.
MULTIMEDIA TOOLS AND APPLICATIONS
(2022)
Article
Genetics & Heredity
Bin Yang, Wenzheng Bao, Jinglong Wang
Summary: In this study, a novel ensemble method based on a flexible neural tree (FNT) was proposed to identify hypertension-related active compounds. The results showed that this ensemble method outperformed other single classifiers in terms of various evaluation metrics, and could more accurately identify hypertension-related compounds compared to classical ensemble methods.
FRONTIERS IN GENETICS
(2021)
Article
Computer Science, Artificial Intelligence
Yanjie Song, Ponnuthurai Nagaratnam Suganthan, Witold Pedrycz, Junwei Ou, Yongming He, Yingwu Chen, Yutong Wu
Summary: This study provides a comprehensive survey on ensemble reinforcement learning (ERL), including its background, motivation, strategies, applications, and future research directions.
APPLIED SOFT COMPUTING
(2023)
Article
Computer Science, Information Systems
Ayse Serbetci, Yusuf Sinan Akgul
Summary: This study proposes a multi-branch ensemble model to address the gradient mismatch problem in Binary Neural Networks (BNN) for efficient person re-identification (ReID). The proposed model outperforms single BNN and conventional ensemble of BNN in terms of performance, computational cost, and memory footprint. The study explores the application of BNNs for efficient person ReID and validates the effectiveness of the proposed ensemble model in image classification.
Article
Computer Science, Artificial Intelligence
Francesca Del Bonifro, Maurizio Gabbrielli, Antonio Lategano, Stefano Zacchiroli
Summary: Image-based programming language identification using convolutional neural networks and transfer learning can accurately recognize a large number of programming languages, with symbols playing the most important role in visual recognizability followed by alphabetic characters.
PEERJ COMPUTER SCIENCE
(2021)
Article
Mathematical & Computational Biology
Ian H. Stevenson
JOURNAL OF COMPUTATIONAL NEUROSCIENCE
(2016)
Article
Biochemical Research Methods
Abed Ghanbari, Aleksey Malyshev, Maxim Volgushev, Ian H. Stevenson
PLOS COMPUTATIONAL BIOLOGY
(2017)
Article
Neurosciences
Ahmad F. Osman, Christopher M. Lee, Monty A. Escabi, Heather L. Read
JOURNAL OF NEUROSCIENCE
(2018)
Article
Biochemical Research Methods
Fatemeh Khatami, Markus Woehr, Heather L. Read, Monty A. Escabi
PLOS COMPUTATIONAL BIOLOGY
(2018)
Article
Computer Science, Artificial Intelligence
Ian H. Stevenson
NEURAL COMPUTATION
(2018)
Article
Biochemistry & Molecular Biology
Chen Chen, Heather L. Read, Monty A. Escabi
Article
Neurosciences
Abed Ghanbari, Naixin Ren, Christian Keine, Carl Stoelzel, Bernhard Englitz, Harvey A. Swadlow, Ian H. Stevenson
JOURNAL OF NEUROSCIENCE
(2020)
Article
Biochemical Research Methods
Fatemeh Khatami, Monty A. Escabi
PLOS COMPUTATIONAL BIOLOGY
(2020)
Article
Neurosciences
Naixin Ren, Shinya Ito, Hadi Hafizi, John M. Beggs, Ian H. Stevenson
JOURNAL OF NEUROPHYSIOLOGY
(2020)
Article
Multidisciplinary Sciences
Xiu Zhai, Fatemeh Khatami, Mina Sadeghi, Fengrong He, Heather L. Read, Ian H. Stevenson, Monty A. Escabi
PROCEEDINGS OF THE NATIONAL ACADEMY OF SCIENCES OF THE UNITED STATES OF AMERICA
(2020)
Article
Biochemical Research Methods
Fengrong He, Ian J. Stevenson, Monty Escabi
Summary: Theories of efficient coding propose that the auditory system is optimized for the statistical structure of natural sounds. This study examines the transformations of peripheral and mid-level auditory filters on the representation of natural sound spectra and modulation statistics. The findings suggest that the tuning characteristics of the peripheral and mid-level auditory system produce a whitened output representation that reduces redundancies and allows for a more efficient use of neural resources.
PLOS COMPUTATIONAL BIOLOGY
(2023)