检索结果-内蒙古大学图书馆

33rd ACM Web Conference, WWW 2024

作者： Ma, Haiping Yang, Yong Qin, Chuan Yu, Xiaoshan Yang, Shangshang Zhang, Xingyi Zhu, Hengshu Information Materials and Intelligent Sensing Laboratory of Anhui Province Institutes of Physical Science and Information Technology Anhui University Anhui Hefei China Institutes of Physical Science and Information Technology Anhui University Anhui Hefei China Career Science Lab BOSS Zhipin PBC School of Finance Tsinghua University Beijing China School of Artificial Intelligence Anhui University Anhui Hefei China The Key Laboratory of Intelligent Computing and Signal Processing of Ministry of Education School of Computer Science and Technology Anhui University Anhui Hefei China Career Science Lab BOSS Zhipin Beijing China

ISBN: (纸本)9798400701719

Knowledge tracing (KT) is a crucial task in online learning, aimed at tracing and predicting each student's knowledge states throughout their learning process. Over the past decade, it has garnered widespread attention due to it provides the potential for more tailored and adaptive online learning experiences. Although most current KT methodologies emphasize optimizing network structures to enhance predictive accuracy for future student performance, they often neglect anomalous interactions in students' learning processes, which may arise from low data quality (i.e., inferior question quality) and abnormal student behaviors (i.e., guessing and mistakes). To this end, in this paper, we propose a novel framework, termed HD-KT, designed to enhance the robustness of existing KT methodologies with Hybrid learning interactions Denoising approach. Specifically, we introduce two detectors for anomalous learning interactions, namely knowledge state-guided anomaly detector and student profile-guided anomaly detector. In the first detection module, we design a sequential autoencoder to identify anomalous learning interactions by detecting atypical student knowledge states. In the second module, we incorporate an attention mechanism by modeling a student's long-term profile to capture irregular interactions. Extensive experiments on four real-world benchmark datasets have decisively shown our HD-KT markedly boosts the robustness of numerous prevailing KT models, consequently increasing the accuracy of future student performance predictions. Additionally, our case studies highlight the versatility of HD-KT in addressing diverse downstream tasks, such as exercise quality analysis and learning behavior-based student clustering. © 2024 ACM.

关键词： Students

来源：评论

学校读者我要写书评

暂无评论

Incipient Fault Detection for Nonlinear Industrial Processes Based on Distributed Weighted Stacked Autoencoder

Incipient Fault Detection for Nonlinear Industrial Processes...

引用

2021 China Automation Congress, CAC 2021

作者： Gao, Huihui Huang, Wenjie Gao, Xuejin Wei, Chen Faculty of Information Technology Beijing University of Technology Engineering Research Center of Digital Community Ministry of Education Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing China

ISBN: (纸本)9781665426473

Incipient faults in industrial processes can deteriorate into abnormal events or even severe accidents. It is still a challenge to detect them accurately due to their slight fault symptoms. In this paper, a distributed weighted stacked autoencoder (DWSAE)-based fault detection scheme is proposed to detect incipient faults by fully utilizing the local and global fault information. First, the local stacked autoencoder models are established for each operation unit, and the feature and residual spaces are obtained. The sliding window technique and singular value decomposition (SVD) are applied to the feature spaces and residual spaces to obtain local fault information indexes. The feature vectors and residual vectors are weighted by the fault information indexes and two local detection statistics in each operation unit are constructed. Second, the global stacked autoencoder model is established using all features extracted by the local stacked autoencoders and two global detection statistics are constructed. Finally, the effectiveness and practicability of the proposed method are verified by a numerical example and Tennessee Eastman process. © 2021 IEEE

关键词： Fault detection

来源：评论

学校读者我要写书评

暂无评论

Oracle Character Recognition using Unsupervised Discriminative Consistency Network

arXiv

引用

arXiv 2023年

作者： Wang, Mei Deng, Weihong Su, Sen The Pattern Recognition and Intelligent System Laboratory School of Artificial Intelligence Beijing University of Posts and Telecommunications Beijing100876 China State Key Laboratory of Networking and Switching Technology Beijing University of Posts and Telecommunications Beijing100876 China

Ancient history relies on the study of ancient characters. However, real-world scanned oracle characters are difficult to collect and annotate, posing a major obstacle for oracle character recognition (OrCR). Besides, serious abrasion and inter-class similarity also make OrCR more challenging. In this paper, we propose a novel unsupervised domain adaptation method for OrCR, which enables to transfer knowledge from labeled handprinted oracle characters to unlabeled scanned data. We leverage pseudo-labeling to incorporate the semantic information into adaptation and constrain augmentation consistency to make the predictions of scanned samples consistent under different perturbations, leading to the model robustness to abrasion, stain and distortion. Simultaneously, an unsupervised transition loss is proposed to learn more discriminative features on the scanned domain by optimizing both between-class and within-class transition probability. Extensive experiments show that our approach achieves state-of-the-art result on Oracle-241 dataset and substantially outperforms the recently proposed structure-texture separation network by 15.1%. © 2023, CC BY.

关键词： Character recognition

来源：评论

学校读者我要写书评

暂无评论

Generalized Predictive Control of Converter Inlet Temperature in the Process of Acid Production with Flue Gas

Generalized Predictive Control of Converter Inlet Temperatur...

引用

Data Driven Control and Learning systems (DDCLS)

作者： Minghua Liu Xiaoli Li Kang Wang Faculty of Information Technology Beijing University of Technology Beijing China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing China Engineering Research Center of Digital Community Ministry of Education Beijing China

The effective control of converter inlet temperature in the process of acid production with flue gas is an effective means to improve the conversion rate of sulfur dioxide and reduce environmental pollution. According to the characteristics of the process of acid production with flue gas, the control process of converter inlet temperature is studied in this paper. Firstly, the CARIMA (Controlled auto-regressive integrated moving average, CARIMA) model of converter inlet temperature is established. Then, a generalized predictive controller based on CARIMA model is designed. Finally, the proposed method is verified by experiment and compared with PID controller. Experimental results show that the proposed method has a better tracking effect and smaller error. The effectiveness of the proposed method is verified.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Flotation Condition Recognition Based on HGNN and Forth Image Dynamic Feature

Flotation Condition Recognition Based on HGNN and Forth Imag...

引用

Data Driven Control and Learning systems (DDCLS)

作者： Zunguan Fan Kang Wang Xiaoli Li Faculty of Information Technology Beijing University of Technology Beijing China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing China Engineering Research Center of Digital Community of Ministry of Education Beijing China

The quality of flotation conditions directly affects the flotation efficiency. Aiming at the problems of difficult online detection, strong subjective arbitrariness, and low recognition efficiency of various flotation conditions in actual flotation work, a flotation condition recognition method based on hypergraph neural network (HGNN) and dynamic feature of forth images is proposed in this paper. Firstly, an improved local binary mode (LBP-TOP) algorithm is introduced to extract the dynamic features of forth sequence containing time information, and then features such as kurtosis and skewness are extracted as supplements to integrate the dynamic features of forth with the supplementary features. By utilizing the aforementioned characteristics and constructing a hypergraph, we have developed an HGNN model that facilitates high-order complex data correlation encoding, thus accomplishing accurate identification of flotation conditions. Finally, simulation shows the effectiveness of the proposed method.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Multi-objective Optimization Control of Flotation Process Based on Policy Iteration

Multi-objective Optimization Control of Flotation Process Ba...

引用

Data Driven Control and Learning systems (DDCLS)

作者： Hang Xu Kang Wang Xiaoli Li Faculty of Information Technology Beijing University of Technology Beijing China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing China Engineering Research Center of Digital Community of Ministry of Education Beijing China

This paper proposes a data-driven, multi-objective optimization control method based on policy iteration for flotation process, addressing the limitations of existing control methods that rely on the system model and cannot satisfy multiple performance indices simultaneously. Firstly, The ordinary linear quadratic regulator algorithm is improved with process data, enabling the algorithm to obtain optimal feedback gain without relying on the internal dynamics model of the system. Then, This method is further extended to the situation where multiple objectives exist, minimizing the deviation of the final additive amount while meeting control requirements for both concentrate and tailings grade. Finally, the convergence and effectiveness of the proposed method are verified by simulation experiments.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Research on a panoramic video transmission scheme based on multi-view switching within the human eye's field of view

Research on a panoramic video transmission scheme based on m...

引用

第34届中国控制与决策会议

作者： Jiangeng Li Kai Xu Guangsheng Wang College of Automation Faculty of Information TechnologyBeijing University of Technology Beijing Key Laboratory of Computational Intelligence and Intelligent System

In today's Internet era,people have increasingly higher requirements for video clarity and breadth of *** scene information that can be expressed by traditional video is limited,and it is difficult to meet people's needs anymore,and panoramic video has emerged as the times *** video can bring people an immersive and interactive audio-visual ***,due to its own characteristics,panoramic video with ultra-high resolution and ultra-high bit rate has a high demand for network bandwidth during network *** paper compares the existing panoramic video transmission schemes,analyzes the advantages and disadvantages of each scheme,and proposes a transmission scheme based on multi-view switching within the human eye' s field of view,and then verifies the effectiveness of the scheme in saving network bandwidth by building a live platform.

关键词： Panoramic Video Viewing Angle Switching Field of View Bandwidth Saving

来源：评论

学校读者我要写书评

暂无评论

TrafficGPT: Viewing, Processing and Interacting with Traffic Foundation Models

arXiv

引用

arXiv 2023年

作者： Zhang, Siyao Fu, Daocheng Zhang, Zhao Yu, Bin Cai, Pinlong School of Transporta-tion Science and Engineering Beihang University Beijing China Key Laboratory of Intelligent Transportation Technology and System Ministry of Education Beijing China Shanghai Artificial Intelligence Laboratory Shanghai China

With the promotion of chatgpt to the public, Large language models indeed showcase remarkable common sense, reasoning, and planning skills, frequently providing insightful guidance. These capabilities hold significant promise for their application in urban traffic management and control. However, LLMs struggle with addressing traffic issues, especially processing numerical data and interacting with simulations, limiting their potential in solving traffic-related challenges. In parallel, specialized traffic foundation models exist but are typically designed for specific tasks with limited input-output interactions. Combining these models with LLMs presents an opportunity to enhance their capacity for tackling complex traffic-related problems and providing insightful suggestions. To bridge this gap, we present TrafficGPT-a fusion of ChatGPT and traffic foundation models. This integration yields the following key enhancements: 1) empowering ChatGPT with the capacity to view, analyze, process traffic data, and provide insightful decision support for urban transportation system management;2) facilitating the intelligent deconstruction of broad and complex tasks and sequential utilization of traffic foundation models for their gradual completion;3) aiding human decision-making in traffic control through natural language dialogues;and 4) enabling interactive feedback and solicitation of revised outcomes. By seamlessly intertwining large language model and traffic expertise, TrafficGPT not only advances traffic management but also offers a novel approach to leveraging AI capabilities in this domain. The TrafficGPT demo can be found in https://***/lijlansg/***. Copyright © 2023, The Authors. All rights reserved.

关键词： Decision making

来源：评论

学校读者我要写书评

暂无评论

Robust Policy Learning Control Design for Multiplayer Nonzero-Sum Games with Uncertainties

Robust Policy Learning Control Design for Multiplayer Nonzer...

引用

Chinese Control Conference (CCC)

作者： Ao Liu Ding Wang Menghua Li Faculty of Information Technology Beijing University of Technology Beijing China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Beijing China Beijing Institute of Artificial Intelligence Beijing University of Technology Beijing China Beijing Laboratory of Smart Environmental Protection Beijing University of Technology Beijing China

This paper investigates robust control problems of nonlinear continuous-time multiplayer systems with infinite horizon by utilizing adaptive dynamic programming algorithms. Combined with pre-training, an improved policy iteration (PI) algorithm is developed to solve robust control issues of multiplayer nonzero-sum (NZS) games with actuator uncertainties. Pre-training of initial weights is added to the PI algorithm to relax the requirement of the initial admissible control policy. It implies that the admissible control pair can be obtained by pre-training of initial weights given randomly. Then, critic neural networks (NNs) are utilized to approximate the optimal control pair by applying the PI algorithm. Robust controllers can be obtained by modifying the optimal control pair. The algorithm accomplishes robust stabilization of multiplayer NZS games with uncertainties. Besides, initial weights of NNs can be set arbitrarily. Finally, a simulation example is given to demonstrate the effectiveness of the developed algorithm.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Generalized Value Iteration for Nonaffine Adaptive Tracking Control with Application to a Wastewater Treatment system 6

Generalized Value Iteration for Nonaffine Adaptive Tracking ...

引用

6th International Conference on Robotics and Automation Engineering, ICRAE 2021

作者： Gao, Ning Wang, Ding Zhao, Mingming Hu, Lingzhi Beijing University of Technology Faculty of Information Technology Beijing China Beijing Key Laboratory of Computational Intelligence and Intelligent System Beijing University of Technology Beijing100124 China Beijing Institute of Artificial Intelligence Beijing University of Technology Beijing100124 China

ISBN: (纸本)9781665406970

In this paper, the generalized value iteration algorithm is utilized to design the tracking controller for the nonaffine wastewater treatment plant. First, the algorithm is initialized by a positive define function to guarantee that the cost function sequence is monotonically nonincreasing. Then, the updating rules of the cost function and the tracking control policy are given under the generalized value iteration framework. Moreover, the model network is constructed to solve the steady control and approximate the system state. The critic and action networks are introduced to approximate the iterative cost function and the iterative tracking control, respectively. Finally, the experimental simulation is provided to validate the feasibility of the present strategy. It is highlighted that the evolving policies during the iterative process are adopted to regulate the concentrations of dissolved oxygen and nitrate nitrogen. © 2021 IEEE.

关键词： Wastewater treatment

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：