检索结果-内蒙古大学图书馆

IDENTIFIABLE LATENT POLYNOMIAL CAUSAL MODELS THROUGH THE LENS OF CHANGE

学校读者我要写书评

暂无评论

arXiv 2023年

作者： Liu, Yuhang Zhang, Zhen Gong, Dong Gong, Mingming Huang, Biwei van den Hengel, Anton Zhang, Kun Shi, Javen Qinfeng Australian Institute for Machine Learning The University of Adelaide Australia School of Computer Science and Engineering The University of New South Wales Australia School of Mathematics and Statistics The University of Melbourne Australia University of California San Diego United States Department of Philosophy Carnegie Mellon University United States Mohamed bin Zayed University of Artificial Intelligence United Arab Emirates

Causal representation learning aims to unveil latent high-level causal representations from observed low-level data. One of its primary tasks is to provide reliable assurance of identifying these latent causal models, known as identifiability. A recent breakthrough explores identifiability by leveraging the change of causal influences among latent causal variables across multiple environments (Liu et al., 2022). However, this progress rests on the assumption that the causal relationships among latent causal variables adhere strictly to linear Gaussian models. In this paper, we extend the scope of latent causal models to involve nonlinear causal relationships, represented by polynomial models, and general noise distributions conforming to the exponential family. Additionally, we investigate the necessity of imposing changes on all causal parameters and present partial identifiability results when part of them remains unchanged. Further, we propose a novel empirical estimation method, grounded in our theoretical finding, that enables learning consistent latent causal representations. Our experimental results, obtained from both synthetic and real-world data, validate our theoretical contributions concerning identifiability and consistency. Copyright © 2023, The Authors. All rights reserved.

关键词： machine learning

Plant Leaf Disease Detection Using EfficientNet Asynchronous Propagation Penalized Neural Network with Planet Optimization Algorithm 2

学校读者我要写书评

暂无评论

Plant Leaf Disease Detection Using EfficientNet Asynchronous...

2nd International Conference on machine learning and Autonomous Systems, ICMLAS 2025

作者： Sasikala, G. Rikhari, Suneetha Sadhana, R. Gavali, V.S. Indra, G. Mamatha, N.P. SRM Valliammai Engineering College Kattankulathur Department of Mathematics Tamil Nadu Chennai603203 India Malla Reddy University Department of Computer Science and Engineering Telangana Hyderabad500100 India Sri Sairam College of Engineering Department of Computer Science and Engineering Karnataka Bengaluru562106 India New Institute of Technology Department of Artificial Intelligence and Machine Learning Maharashtra Kolhapur416005 India R M K College of Engineering and Technology Department of Computer Science and Engineering Tamil Nadu Puduvoyal601206 India Brindavan College of Engineering Department of Electronics and Communication Engineering Karnataka Bangalore560063 India

ISBN: (纸本)9798331505745

Unpredictability in illumination conditions, similar illness symptoms, skewed datasets, processing expense, and the challenge of determining infections at their onset are some of the challenges in plant leaf disease diagnosis. Model accuracy, reliability, and generalization across different types of plant species are also influenced by environmental factors, occlusions, and image noise. Using PlantVillage database, this paper confirms the superiority of the EfficientNet Asynchronous Propagation Penalized Neural Network with Planet Optimization Algorithm (ENet-APPNNet-POA) in plant leaf disease detection. Image quality is initially enhanced by the Adaptive Tri-Plateau Limit Tri-Histogram Algorithm (ATP-LTH) approach. ENet-UNet (EfficientNet-based U-Net) divides the affected areas of plant leaves according to distinct regions. Asynchronous Propagation Penalized Neural Network (APPNNet) serves as the classifier responsible for extracting features before performing their category identification. The Planet Optimization Algorithm (POA) delivers improved results and accuracy in classifying plant leaf diseases by evaluating various classifications affecting different leaf areas. PlantVillage dataset is employed in the Python test script. Test results indicate that ENet-APPNNet-POA outperforms existing methods to detect plant leaf disease classes at 99.9% efficiency and 99.8% sensitivity, an indication that computerized systems are likely to supersede manual diagnosis. © 2025 IEEE.

关键词： Image enhancement

A Reinforcement learning Method in Cooperative Multi-Agent System for Production Control System

学校读者我要写书评

暂无评论

A Reinforcement Learning Method in Cooperative Multi-Agent S...

Data Science and Network Security (ICDSNS), International Conference on

作者： V. Malathy Hassan M. Al-Jawahry Madhura G K G. Suganya Rashmi P Department of ECE SR University Waranga India Department of Computers Techniques Engineering College of Technical Engineering The Islamic university Najaf Iraq Department of Artificial intelligence and Machine learning Nitte Meenakshi Institute of Technology Bengaluru India Department of Computer Science Dr. SNS Rajalakshmi College of Arts and Science Coimbatore India Inventuriz Labs Private Ltd Bengaluru India

ISBN: (数字)9798350373110

ISBN: (纸本)9798350373127

Nowadays, in various domains including distributed control, telecommunication, robotics and economics to address the problems, Multi-Agent Systems (MAS) is used. To solve with preprogrammed agent behavior complexity of many tasks of these domains making it difficult. The multi- agent systems can limit problem of control in complex system production, to solve more efficiently. However, local optimization tendencies are often shown in iterative development algorithms. So, by using reinforcement learning and considering indicators of global key performance this paper presents a new method for cooperative multi-agent system. To cooperative order agents, for the purpose of this a central deep and double deep learning module its knowledge is being transferred. From the results, subsequent reinforcement learning using memory, the order's experience is stored and increases 50% of average mean cycle. In comparison of production control conventional methods, nonlinear production system and parameters of learning interdependencies between the characteristics are investigated and the performance is evaluated.

关键词： Production systems Job shop scheduling Reinforcement learning Production control Network security Complexity theory Manufacturing Optimization Robots Multi-agent systems

BrainIB: Interpretable Brain Network-based Psychiatric Diagnosis with Graph Information Bottleneck

学校读者我要写书评

暂无评论

arXiv 2022年

作者： Zheng, Kaizhong Yu, Shujian Li, Baojuan Jenssen, Robert Chen, Badong National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an China The Department of Computer Science Vrije Universiteit Amsterdam Amsterdam and the Machine Learning Group UiT - Arctic University of Norway Tromsø Norway The Machine Learning Group UiT - Arctic University of Norway Tromsø Norway The School of Biomedical Engineering Fourth Military Medical University Xi’an China

Developing a new diagnostic models based on the underlying biological mechanisms rather than subjective symptoms for psychiatric disorders is an emerging consensus. Recently, machine learning-based classifiers using functional connectivity (FC) for psychiatric disorders and healthy controls are developed to identify brain markers. However, existing machine learning-based diagnostic models are prone to over-fitting (due to insufficient training samples) and perform poorly in new test environment. Furthermore, it is difficult to obtain explainable and reliable brain biomarkers elucidating the underlying diagnostic decisions. These issues hinder their possible clinical applications. In this work, we propose BrainIB, a new graph neural network (GNN) framework to analyze functional magnetic resonance images (fMRI), by leveraging the famed Information Bottleneck (IB) principle. BrainIB is able to identify the most informative edges in the brain (i.e., subgraph) and generalizes well to unseen data. We evaluate the performance of BrainIB against 3 baselines and 7 state-of-the-art brain network classification methods on three psychiatric datasets and observe that our BrainIB always achieves the highest diagnosis accuracy. It also discovers the subgraph biomarkers which are consistent to clinical and neuroimaging findings. The source code and implementation details of BrainIB are freely available at GitHub repository (https://***/SJYuCNEL/brain-and-Information-Bottleneck/). Copyright © 2022, The Authors. All rights reserved.

关键词： Graph neural networks

A human activity recognition approach based on skeleton extraction and image reconstruction 21

学校读者我要写书评

暂无评论

A human activity recognition approach based on skeleton extr...

5th International Conference on Graphics and Signal Processing, ICGSP 2021

作者： Chen, Yanbing Ke, Wei Chan, Ka-Hou Xiong, Zhang School of Applied Sciences Macao Polytechnic Institute Macao SAR China Engineering Research Centre of Applied Technology on Machine Translation and Artificial Intelligence Ministry of Education Macao Polytechnic Institute Macao SAR China School of Computer Science and Engineering Beihang University Beijing China

ISBN: (纸本)9781450389419

Human Activity Recognition (HAR) has wide applications in surveillance, assisted living and robotic systems that interact with people. Traditional methods for HAR usually input source images directly. Various factors of these direct source images can affect the accuracy of recognition, such as heights, weights, poses, angles and whether being obscured or not. Most of the latest research on HAR has mainly focused on the use of 2D skeleton data, extracted from videos captured by standard cameras. Inspired by this trend, we propose a novel way of using skeleton data in this paper. As a pre-processing phase, we adopt the OpenPose library to extract 2D positions of human skeleton joints, then reconstruct, or synthesize, the skeleton image from these joints, together with coloring to encode the categories of different human parts. The reconstructed images are put into a Convolutional Neural Network (CNN) structure for classification, to obtain the recognition result. We implement the typical four activity classes, i.e., squat, stand, walk and work, in our experiments. Compared with feeding the original complex source images, our approach reduces the complexity of the neural network thus the time consumption, and improves the recognition accuracy significantly. All the images that we used for training and testing are collected from the real public places. We achieved a recognition accuracy of 97.3% in our experiments, indicating that our method has very good performance. Thanks to the lightning speed of the OpenPose library, our method can be carried out in real-time. The demonstration of using synthesized input to bridge existing technologies also has its general significance. © 2021 ACM.

关键词： Convolutional neural networks

Development of a Calibration-Free Brain-computer Interface Utilizing Common Spatial Patterns and artificial Neural Networks for EEG Signal Analysis

学校读者我要写书评

暂无评论

Development of a Calibration-Free Brain-Computer Interface U...

Human-computer Interaction, Optimization and Robotic Applications (HORA), International Congress on

作者： Korhan Cengiz Shreyas J Udayaprasad P K Gururaj H L Dilip Kumar S M College of Information Technology University of Fujairah UAE Faculty of Organization and Informatics University of Zagreb Croatia Department of Information Technology Manipal Institute of Technology Bengaluru Manipal Academy of Higher Education Manipal Karnataka India Department of Artificial Intelligence & Machine Learning Global Academy of Technology Bengaluru India Dept. of Computer Science and Engineering University Visvesvaraya College of Engineering Bangalore India

ISBN: (数字)9798350394634

ISBN: (纸本)9798350394641

A brain computer interface (BCI) system uses a technique known calibration, that takes 20 to 30 minutes to accomplish. For the objective of creating a reliable decoder, the calibration process is challenging and expensive. In order to address the drawbacks of the current system, a spectral-spatial technique has been suggested. The motor imagery (MI) data set, comprising 15 electroencephalography (EEG) signals and fourteen test subjects, is taken into consideration. The two modules are designed to extract characteristics and process data. An artificial neural network (ANN) is used to independently train and test the suggested spectral-spatial algorithm. Based on it, a variety of machine learning techniques, including random forest (RF), neural networks (NN), and XGboost, are used to classify the data, that is then sent to the hidden layer (Lth layer). The obtained results indicates 2% of improvement in comparison with existing methodology.

关键词： Training machine learning algorithms artificial neural networks Electroencephalography Brain-computer interfaces Calibration Reliability

Perceptual Enhancement Via Model Compression: Introducing the Eloss Metric

学校读者我要写书评

暂无评论

SSRN

SSRN 2024年

作者： Li, Chuangqi Liu, Ziyu Li, Fukun Liu, Liuxin Yang, Yifan Che, Gaofeng College of Computer and Information Engineering Henan Normal University Henan Xinxiang China Henan Key Laboratory of Educational Artificial Intelligence and Personalized Learning China The School of Vehicle and Mobility Tsinghua University Beijing China Northeast Normal University Jilin Changchun China The University of Sydney Sydney Australia

The significance of safety perception in intelligent driving systems has increased as traffic environments become more complex. In this paper, we propose a novel measurement method called Eloss and a complementary training strategy to enhance the capability of perception models in handling abnormal data in complex traffic scenarios. Addressing the limitations of traditional robust perception methods in real-world environments, Eloss provides a comprehensive analysis of the information compression layer of perception models. Inspired by communication system design principles, Eloss ensures stable information variation and continuous reduction in information entropy. By guiding network parameter updates, Eloss generates highly sensitive measurement standards for accurately detecting abnormal data while maintaining model performance stability. Experimental results demonstrate that Eloss exhibits significant deviations when encountering abnormal data and can differentiate between different yet similar abnormalities, validating its effectiveness and practicality. This approach robustly supports enhancing safety perception levels in intelligent driving systems. © 2024, The Authors. All rights reserved.

关键词： Data handling

ERetinex: Event Camera Meets Retinex Theory for Low-Light Image Enhancement

学校读者我要写书评

暂无评论

arXiv 2025年

作者： Guo, Xuejian Tian, Zhiqiang Wang, Yuehang Li, Siqi Jiang, Yu Du, Shaoyi Gao, Yue the School of Software Engineering Xi’an Jiaotong University Xi’an710049 China the College of Computer Science and Technology Jilin University Changchun130012 China the National Key Laboratory of Human-Machine Hybrid Augmented Intelligence National Engineering Research Center for Visual Information and Applications Institute of Artificial Intelligence and Robotics Xi’an Jiaotong University Xi’an710049 China BNRist THUIBCS KLISS BLBCI School of Software Tsinghua University Beijing100084 China

Low-light image enhancement aims to restore the under-exposure image captured in dark scenarios. Under such scenarios, traditional frame-based cameras may fail to capture the structure and color information due to the exposure time limitation. Event cameras are bio-inspired vision sensors that respond to pixel-wise brightness changes asynchronously. Event cameras’ high dynamic range is pivotal for visual perception in extreme low-light scenarios, surpassing traditional cameras and enabling applications in challenging dark environments. In this paper, inspired by the success of the retinex theory for traditional frame-based low-light image restoration, we introduce the first methods that combine the retinex theory with event cameras and propose a novel retinex-based low-light image restoration framework named ERetinex. Among our contributions, the first is developing a new approach that leverages the high temporal resolution data from event cameras with traditional image information to estimate scene illumination accurately. This method outperforms traditional image-only techniques, especially in low-light environments, by providing more precise lighting information. Additionally, we propose an effective fusion strategy that combines the high dynamic range data from event cameras with the color information of traditional images to enhance image quality. Through this fusion, we can generate clearer and more detail-rich images, maintaining the integrity of visual information even under extreme lighting conditions. The experimental results indicate that our proposed method outperforms state-of-the-art (SOTA) methods, achieving a gain of 1.0613 dB in PSNR while reducing FLOPS by 84.28%. The code is available at https://***/lodew920/ERetinex. © 2025, CC BY.

关键词： Cameras

Contrastive Continual learning with Importance Sampling and Prototype-Instance Relation Distillation

学校读者我要写书评

暂无评论

arXiv 2024年

作者： Li, Jiyong Azizov, Dilshod Li, Yang Liang, Shangsong School of Computer Science and Engineering Sun Yat-sen University China Guangdong Key Laboratory of Big Data Analysis and Processing Guangzhou China Department of Machine Learning Mohamed bin Zayed University of Artificial Intelligence United Arab Emirates AI Thrust Information Hub The Hong Kong University of Science and Technology Guangzhou China Department of CSE The Hong Kong University of Science and Technology China

Recently, because of the high-quality representations of contrastive learning methods, rehearsal-based contrastive continual learning has been proposed to explore how to continually learn transferable representation embeddings to avoid the catastrophic forgetting issue in traditional continual settings. Based on this framework, we propose Contrastive Continual learning via Importance Sampling (CCLIS) to preserve knowledge by recovering previous data distributions with a new strategy for Replay Buffer Selection (RBS), which minimize estimated variance to save hard negative samples for representation learning with high quality. Furthermore, we present the Prototype-instance Relation Distillation (PRD) loss, a technique designed to maintain the relationship between prototypes and sample representations using a self-distillation process. Experiments on standard continual learning benchmarks reveal that our method notably outperforms existing baselines in terms of knowledge preservation and thereby effectively counteracts catastrophic forgetting in online contexts. The code is available at https://***/lijy373/CCLIS. Copyright © 2024, The Authors. All rights reserved.

关键词： Distillation