检索结果-内蒙古大学图书馆

Mango wine making process optimization based on artificial intelligence deep learning technology

EXPERT SYSTEMS 2024年第6期41卷 e13032-e13032页

作者： Hua Xubin Lin Qiao Gong Fayong Cai Li Liu Junhua Xichang Univ Xichang 615000 Sichuan Peoples R China Panxi Crops Res & Utilizat Key Lab Sichuan Prov Xichang Sichuan Peoples R China Liangshan Tianmeiyi Agr Sci & Technol Co Ltd Xichang Sichuan Peoples R China

This paper combines artificial intelligence deep learning technology to optimize the wine making process of mango wine. Moreover, in view of the shortcomings of traditional electronic nose data processing methods, a deep learning method based on SSAE-BPNN is proposed for electronic nose data processing. In addition, according to the characteristics of automatic learning features, this paper uses a deep learning method based on SSAE-BPNN to simplify the process of traditional data processing methods. Finally, this paper constructs an electronic nose system that can be used to identify mango wine making characteristics, and enhances the effect of electronic nose recognition through deep learning. Through the analysis, it can be seen that the mango wine making process optimization method based on artificial intelligence deep learning technology proposed in this paper has a certain effect, and it has optimized the traditional mango wine making process.

关键词： auto-encoder deep learning electronic nose system optimization SSAE-BPNN

来源：评论

学校读者我要写书评

暂无评论

Advancing spatiotemporal forecasts of CO2 plume migration using deep learning networks with transfer learning and interpretation analysis

引用

INTERNATIONAL JOURNAL OF GREENHOUSE GAS CONTROL 2024年 132卷

作者： Fan, Ming Wang, Hongsheng Zhang, Jing Hosseini, Seyyed A. Lu, Dan Oak Ridge Natl Lab Computat Sci & Engn Div Oak Ridge TN 38831 USA Univ Texas Austin Bur Econ Geol Austin TX 78758 USA Emory Univ Dept Comp Sci Atlanta GA 30322 USA

Accurate and timely forecasts of CO2 plume distribution throughout the injection and post -injection phases are crucial for detecting plume migration, assessing leakage risks, and supporting operational decisions in geologic carbon storage (GCS). Current convolutional neural network -based approaches primarily focus on spatial information and overlook temporal dependencies in plume distributions, thus limiting their ability to capture dynamic movement effects and provide accurate predictions of plume migration. In this work, we propose two deep learning models, auto -encoder (AE)-LSTM and encoder -Decoder (ED)-ConvLSTM, each uniquely designed to capture both spatial and temporal features. We apply the proposed methods to forecast the dynamic distribution of CO2 plumes based on 108 reservoir simulations over a 30 -year injection and a 30year post -injection period. The results indicate that the ED-ConvLSTM model outperforms the AE-LSTM model in accurately predicting the spatiotemporal dynamics of CO2 plume migration, achieving R2 values above 0.99. To provide a deeper understanding of these model predictions, we employ a gradient -based explanation method on the trained models. This approach provides insights into the influence of input variables on plume migration forecasts and uncovers the underlying prediction mechanisms of the proposed models. Furthermore, we introduce a transfer learning technique, enabling fast and accurate plume migration forecasting in the post -injection phase by leveraging the trained model during the injection phase. This reduces the necessity for extensive data collection or re-training. The methods proposed in our work enhances the performance and interpretability of CO2 plume migration forecasts, thereby facilitating informed decision -making throughout the entire lifecycle of GCS applications.

关键词： encoder-Decoder ConvLSTM networks auto-encoder Explainable machine learning Transfer learning Geologic carbon storage

来源：评论

学校读者我要写书评

暂无评论

A reusable decoder network penalized by smooth group lasso and its applications to large-scale fault diagnosis of machinery

引用

CONTROL ENGINEERING PRACTICE 2024年 153卷

作者： Zhang, Zhiqiang He, Hongji Xu, Shuiqing Yin, Lisheng Dong, Xueping Hefei Univ Technol Sch Elect Engn & Automat Hefei 230009 Peoples R China

Representation learning approaches have achieved great success in fault diagnosis of large-scale mechanical data, among which the popular auto-encoder method has developed a series of effective variants. In the existing variants, the encoder network is re-employed to encode feature representations of the data, while the decoder network is directly discarded after training, leading to a regrettable waste of computational resources. Instead of proposing advanced variants of the auto-encoder, this paper explicitly penalizes the decoder network with group lasso, thereby transforming waste into treasure. Specifically, the group lasso constrains the column vectors of the decoder network's weight matrix at the group level, making them reusable for feature selection. Moreover, a smooth function is utilized to approximate the group lasso to prevent numerical oscillations when computing the gradients. The simulated data and experimental gear data are sequentially used to verify the effectiveness of the smooth group lasso through investigations on two representative auto-encoder variants. The results show that the decoder network penalized by smooth group lasso can be re-utilized to guide selection of a subset of key features for training a classifier, exhibiting an extraordinary feature selection capability.

关键词： Representation learning auto-encoder Large-scale fault diagnosis Smooth group lasso Feature selection

来源：评论

学校读者我要写书评

暂无评论

Extracting visual features for personalized recommendation using autoencoder

引用

Procedia Computer Science 2024年 246卷 3634-3643页

作者： Aymen Ben Hassen Sonia Ben Ticha Anja Habacha Chaibi RIADI Laboratory National School of Computer Science University of Manouba Tunisia

In recent years, product images have garnered increasing interest within recommendation systems, as the visual appearance of products strongly influences users’ decision. The primary goal of personalized recommendation systems is to provide suggestions that reflect with the individual preferences of each user. Recently, deep learning models have demonstrated remarkable performance, revealing a strong potential in analyzing visual features. This paper present an innovative approach that utilizes item images to construct a personalized model for users. The approach involves employing autoencoder for feature extraction of visual features in item images. Subsequently, these latent features are correlated with user preferences to establish the personalized model, which is then integrated into a collaborative filtering (CF) algorithm for recommendation purposes. The effectiveness of the proposed approach is assessed through experiments on two substantial real datasets, encompassing fashion items from *** and movies from the MovieLens dataset. A comparative analysis with alternative collaborative filtering-based methods is conducted to validate our results.

关键词： Visual features Deep Learning auto-encoder Features Extraction Dimensionality reduction Recommender Systems Personalized User Modeling

来源：评论

学校读者我要写书评

暂无评论

Deep learning based decoding of single local field potential events

引用

NEUROIMAGE 2024年 297卷 120696-120696页

作者： Schilling, Achim Gerum, Richard Boehm, Claudia Rasheed, Jwan Metzner, Claus Maier, Andreas Reindl, Caroline Hamer, Hajo Krauss, Patrick Univ Hosp Erlangen Neurosci Lab Erlangen Germany Univ Erlangen Nurnberg Cognit Computat Neurosci Grp Erlangen Germany York Univ Dept Phys Toronto ON Canada York Univ Ctr Vis Res Toronto ON Canada Univ Erlangen Nurnberg Pattern Recognit Lab Nurnberg Germany Univ Hosp Erlangen Epilepsy Ctr Dept Neurol Erlangen Germany

How is information processed in the cerebral cortex? In most cases, recorded brain activity is averaged over many (stimulus) repetitions, which erases the fine -structure of the neural signal. However, the brain is obviously a single -trial processor. Thus, we here demonstrate that an unsupervised machine learning approach can be used to extract meaningful information from electro-physiological recordings on a single -trial basis. We use an auto -encoder network to reduce the dimensions of single local field potential (LFP) events to create interpretable clusters of different neural activity patterns. Strikingly, certain LFP shapes correspond to latency differences in different recording channels. Hence, LFP shapes can be used to determine the direction of information flux in the cerebral cortex. Furthermore, after clustering, we decoded the cluster centroids to reverse-engineer the underlying prototypical LFP event shapes. To evaluate our approach, we applied it to both extra -cellular neural recordings in rodents, and intra-cranial EEG recordings in humans. Finally, we find that single channel LFP event shapes during spontaneous activity sample from the realm of possible stimulus evoked event shapes. A finding which so far has only been demonstrated for multi-channel population coding.

关键词： Deep learning Local field potentials Auditory neuroscience Auditory cortex Speech perception auto-encoder Embeddings Intracranial EEG (iEEG) Stereotactic EEG (sEEG)

来源：评论

学校读者我要写书评

暂无评论

auto-encoder USING THE BI-FIRING ACTIVATION FUNCTION 13

AUTO-ENCODER USING THE BI-FIRING ACTIVATION FUNCTION

引用

International Conference on Machine Learning and Cybernetics

作者： Cao, Zihong Zeng, Guangjun Ng, Wing W. Y. Le, Jincheng South China Univ Technol Sch Comp Sci & Engn Machine Learning & Cybernet Res Ctr Guangzhou 510006 Guangdong Peoples R China Guangdong Pharmaceut Univ Coll Med Informat Engn Guangzhou 510006 Guangdong Peoples R China South China Univ Technol Coll Comp Sci & Engn Guangzhou 510006 Guangdong Peoples R China

ISBN: (纸本)9781479942152

Training the whole deep neural network together is restricted by the gradient diffusion problem. Greedy layer-wise training of an auto-encoder has achieved promising results in deep neural networks. However, it can not learn useful input representation from the original input directly. In this work, we propose to use the bi-firing activation function for auto-encoder with an end-to-end training scheme. It not only improves the training efficiency but also learns better features than the traditional stacked auto-encoder. Experimental results show that it extracts more representative features and also outperforms the stacked auto-encoder in supervised classification task.

关键词： auto-encoder Deep Learning Bi-firing function Layer-wise scheme

来源：评论

学校读者我要写书评

暂无评论

automatic deep spare clustering with a dynamic population-based evolutionary algorithm using reinforcement learning and transfer learning

引用

IMAGE AND VISION COMPUTING 2024年 151卷

作者： Hadikhani, Parham Lai, Daphne Teck Ching Ong, Wee-Hong Nadimi-Shahraki, Mohammad H. Univ Brunei Darussalam Sch Digital Sci Darussalam Brunei Islamic Azad Univ Fac Comp Engn Najafabad Branch Najafabad Iran

Clustering data effectively remains a significant challenge in machine learning, particularly when the optimal number of clusters is unknown. Traditional deep clustering methods often struggle with balancing local and global search, leading to premature convergence and inefficiency. To address these issues, we introduce ADSCDPE-RT (automatic Deep Sparse Clustering with a Dynamic Population-based Evolutionary Algorithm using Reinforcement Learning and Transfer Learning), a novel deep clustering approach. ADSC-DPE-RT builds on Multi-Trial Vector-based Differential Evolution (MTDE), an algorithm that integrates sparse auto-encoding and manifold learning to enable automatic clustering without prior knowledge of cluster count. However, MTDE's fixed population size can lead to either prolonged computation or premature convergence. Our approach introduces a dynamic population generation technique guided by Reinforcement Learning (RL) and Markov Decision Process (MDP) principles. This allows for flexible adjustment of population size, preventing premature convergence and reducing computation time. Additionally, we incorporate Generative Adversarial Networks (GANs) to facilitate dynamic knowledge transfer between MTDE strategies, enhancing diversity and accelerating convergence towards the global optimum. This is the first work to address the dynamic population issue in deep clustering through RL, combined with Transfer Learning to optimize evolutionary algorithms. Our results demonstrate significant improvements in clustering performance, positioning ADSC-DPE-RT as a competitive alternative to state-of-the-art deep clustering methods.

关键词： Unsupervised learning Deep clustering Reinforcement learning Transfer learning Evolutionary algorithm Differential evolution auto-encoder Dimension reduction

来源：评论

学校读者我要写书评

暂无评论

Dynamic Resource Allocation for Multi-User Goal-oriented Communications at the Wireless Edge 30

Dynamic Resource Allocation for Multi-User Goal-oriented Com...

引用

30th European Signal Processing Conference (EUSIPCO)

作者： Binucci, Francesco Banelli, Paolo Di Lorenzo, Paolo Barbarossa, Sergio Univ Perugia Dept Engn Via G Duranti 93 I-06128 Perugia Italy Sapienza Univ Rome DIET Dept Via Eudossiana 18 I-00184 Rome Italy

ISBN: (纸本)9789082797091

This paper proposes a wireless, goal-oriented, multi-user communication system assisted by edge-computing, within the general framework of Edge Machine Learning (EML). Specifically, we consider a set of mobile devices that, exploiting convolutional encoders (CE), namely the encoder part of the convolutional auto-encoders (CAE), send compressed data units to an edge server (ES) that performs a specific learning task, such as image classification. The training of both the CEs and the ES classification networks is performed in a off-line fashion, employing a cross-entropy loss, regularized by the mean squared error of the CAE expanded output. Then, exploiting such goal-oriented architecture, and employing a Lyapunov optimization framework, we considered the joint management of computation and transmission resources for the overall system. In particular, we considered a Multi-User Minimum Energy Resource Allocation Strategy (mu-MERAS), which provides the optimal resource allocation for both the devices and the ES, in a energy-efficient perspective. Simulation results highlight a classical EML trade-off between energy, latency, and accuracy, as well as the effectiveness of the proposed approach to adaptively manage resources according to wireless channels conditions, computing requests, and classification reliability.

关键词： Goal-Oriented communications auto-encoder Lyapunov Optimization Edge Machine Learning

来源：评论

学校读者我要写书评

暂无评论

Servo Health Monitoring Based on Feature Learning via Deep Neural Network

引用

IEEE ACCESS 2021年 9卷 160887-160896页

作者： Zhou, Yajing Zheng, Yuemin Tao, Jin Sun, Mingwei Sun, Qinglin Dehmer, Matthias Chen, Zengqiang Nankai Univ Coll Artificial Intelligence Tianjin 300350 Peoples R China Aalto Univ Dept Elect Engn & Automat Espoo 02150 Finland Swiss Distance Univ Appl Sci Dept Comp Sci CH-3900 Brig Switzerland

As the core actuator of an aircraft's flight control system, the servos' reliability directly affects the safety of the flight control system and the whole aircraft. The failure of the rudder will lead to the poor control effect of aircraft, affect its flight quality and safety, and even cause major flight accidents. In order to monitor the health status of servo and determine the fault and its degree accurately, this paper presents a feature learning based health monitoring method using a deep neural network. Firstly, we combine the wavelet packet decomposition and support vector machine to synthesize the sample segment label. And then, the sliding window is employed to enlarge the sample size, and the auto-encoder is utilized to reduce the data dimension. Moreover, the Softmax classifier is used for health monitoring. At last, the numerical simulations demonstrate the effectiveness of the proposed method.

关键词： Gears Actuators Monitoring Feature extraction Servomotors Fault diagnosis Support vector machines Servo health wavelet packet decomposition auto-encoder softmax classifier health monitoring

来源：评论

学校读者我要写书评

暂无评论

Pseudo-Supervised Deep Subspace Clustering

引用

IEEE TRANSACTIONS ON IMAGE PROCESSING 2021年 30卷 5252-5263页

作者： Lv, Juncheng Kang, Zhao Lu, Xiao Xu, Zenglin Univ Elect Sci & Technol China Sch Comp Sci & Engn Chengdu 611731 Peoples R China Harbin Inst Technol Dept Comp Sci & Technol Shenzhen 518055 Peoples R China

auto-encoder (AE)-based deep subspace clustering (DSC) methods have achieved impressive performance due to the powerful representation extracted using deep neural networks while prioritizing categorical separability. However, self-reconstruction loss of an AE ignores rich useful relation information and might lead to indiscriminative representation, which inevitably degrades the clustering performance. It is also challenging to learn high-level similarity without feeding semantic labels. Another unsolved problem facing DSC is the huge memory cost due to n x n similarity matrix, which is incurred by the self-expression layer between an encoder and decoder. To tackle these problems, we use pairwise similarity to weigh the reconstruction loss to capture local structure information, while a similarity is learned by the self-expression layer. Pseudo-graphs and pseudo-labels, which allow benefiting from uncertain knowledge acquired during network training, are further employed to supervise similarity learning. Joint learning and iterative training facilitate to obtain an overall optimal solution. Extensive experiments on benchmark datasets demonstrate the superiority of our approach. By combining with the k -nearest neighbors algorithm, we further show that our method can address the large-scale and out-of-sample problems. The source code of our method is available: https://***/sckangz/SelfsupervisedSC.

关键词： Subspace clustering auto-encoder pseudo-graph pseudo-label similarity learning self-expression pseudo-supervision

来源：评论

学校读者我要写书评

暂无评论

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案：

请选择收藏分类：

通借通还

建议与咨询 留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

时间限定

文献类型

馆藏选择

核心期刊

语言

文献类型

帮助

文字说明：

检索规则说明：

检索范例：

分类表

所选分类

限定检索结果

文献类型

馆藏范围

日期分布

学科分类号

主题

机构

作者

语言

请选择保存的检索档案： 新增检索档案 确定 取消

请选择收藏分类： 新增自定义分类 确定 取消

通借通还

建议与咨询留下您的常用邮箱和电话号码，以便我们向您反馈解决方案和替代方法

请选择保存的检索档案：

请选择收藏分类：