检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Su, Jiayu Knowles, David A. Rabadan, Raul Program for Mathematical Genomics United States Department of Systems Biology Columbia University United States Department of Biomedical Informatics Columbia University United States Department of Computer Science Columbia University United States New York Genome Center United States

The success of machine learning models relies heavily on effectively representing high-dimensional data. However, ensuring data representations capture human-understandable concepts remains difficult, often requiring the incorporation of prior knowledge and decomposition of data into multiple subspaces. Traditional linear methods fall short in modeling more than one space, while more expressive deep learning approaches lack interpretability. Here, we introduce Supervised Independent Subspace Principal Component Analysis (sisPCA), a PCA extension designed for multi-subspace learning. Leveraging the Hilbert-Schmidt Independence Criterion (HSIC), sisPCA incorporates supervision and simultaneously ensures subspace disentanglement. We demonstrate sisPCA's connections with autoencoders and regularized linear regression and showcase its ability to identify and separate hidden data structures through extensive applications, including breast cancer diagnosis from image features, learning aging-associated DNA methylation changes, and single-cell analysis of malaria infection. Our results reveal distinct functional pathways associated with malaria colonization, underscoring the essentiality of explainable representation in high-dimensional data analysis. Copyright © 2024, The Authors. All rights reserved.

关键词： Self-supervised learning

来源：评论

学校读者我要写书评

暂无评论

PWM Signal Reconstruction to Operate the iBooster Brake for Electric Vehicle: A Reverse Engineering

PWM Signal Reconstruction to Operate the iBooster Brake for ...

引用

Smart Mechatronics (ICSMech), IEEE International Conference on

作者： Muhamad Fajar Imanul Haq Faiza Latifah Mulia Pratama Dewi Indriati Hadi Putri Muhammad Rizalul Wahid Bambang Wahono Achmad Praptijanto Yanuandri Putrasari Ahmad Dimyani Arifin Nur Muhammad Khristamto Aditya Wardana Suherman Program Study of Mechatronics and Artificial Intelligence Universitas Pendidikan Indonesia Purwakarta Indonesia Research Center for Smart Mechatronics Research Organization of Electronics and Informatics National Research and Innovation Agency Bandung Indonesia Mechatronics and Artificial Intelligence Universitas Pendidikan Indonesia Purwakarta Indonesia

ISBN: (数字)9798350368918

ISBN: (纸本)9798350368925

The rise of Advanced Driver Assistance Systems (ADAS) has emphasized the need for technologies such as Automatic Emergency Braking (AEB) and Adaptive Cruise Control, both of which require brake activation without driver input. Since 2016, automotive manufacturers, particularly in the U.S., have been incorporating AEB technologies in response to safety demands. Bosch's iBooster, an electromechanical brake booster, has gained prominence for its role in autopilot and braking systems, notably in Tesla's Model S and X. The iBooster system can generate full braking pressure in just 120 milliseconds and allows for electronic control while still maintaining manual override capabilities. Tesla's iBooster Gen1, used in its autopilot system, operates via a Controlled Area Network (CAN), though the exact method is not publicly available. Honda has also adopted the iBooster Gen1 in the 2018 CR-V, utilizing it for the Brake Assist feature, although details on Honda's CAN network usage remain limited. This paper explores reverse engineering of the iBooster to uncover key hardware and software technologies. The reverse engineering operation entails altering the system's control loop, capturing the pedal sensor signal, and replicating this signal using an Arduino. Additionally, it includes designing circuit schematics and PCBs. The results demonstrate the success of this reverse engineering approach, which uses a work cycle and Pulse Width Modulation (PWM) to interact with the iBooster Gen1. The method produced identical responses to the iBooster with PWM values set at 30%, 40%, and 50%.

关键词： Reverse engineering Pulse width modulation Software Signal reconstruction Autopilot Safety Brakes Tuning Optimization Vehicles

来源：评论

学校读者我要写书评

暂无评论

Sentiment Analysis of Vaccine Perceptions in Indonesia: A Comparative Study Using the Naïve Bayes Multinomial Text Algorithm on Twitter Data

Sentiment Analysis of Vaccine Perceptions in Indonesia: A Co...

引用

International Seminar on Intelligent Technology and Its Applications (ISITIA)

作者： Dedy Hermawan Addin Aditya Febry Eka Purwiantono Dept. of Informatics Sekolah Tinggi Informatika & Komputer Indonesia Malang Indonesia Dept. of Information System Sekolah Tinggi Informatika & Komputer Indonesia Malang Indonesia Diploma Program of Information System Sekolah Tinggi Informatika & Komputer Indonesia Malang Indonesia

ISBN: (数字)9798350378573

ISBN: (纸本)9798350378580

In the ongoing COVID-19 situation, vaccines have played a main role in mitigating the propagation of the virus and reducing its impact on public health. In Indonesia, various vaccines have been deployed, including Sinovac, AstraZeneca, Pfizer, and Moderna. Understanding public perception of these vaccines is vital for addressing vaccine hesitancy and ensuring widespread immunization coverage. Public sentiment can significantly influence vaccination uptake, making it essential to analyze and understand these perceptions accurately. This study focused on analyzing public perception regarding the use of the Sinovac, AstraZeneca, Pfizer, and Moderna vaccines in Indonesia. Classifying tweets using The Naïve Bayes algorithm is help to compare either positive or negative sentiments. Data collection is conducted using the Twitter API (Application programming Interface). The data analysis process involves grouping data and assigning labels, followed by preprocessing steps including case folding, tokenizing, filtering tokens by length, removing stop words, and stemming. The Naïve Bayes Multinomial Text algorithm is chosen for its capability to classify documents based on word occurrence frequencies. Results from the processed data show the accuracy rates for the Sinovac vaccine category with a dataset of 5000 tweets at 95%, AstraZeneca vaccine with a dataset of 1500 tweets at 98.01%, Pfizer vaccine with a dataset of 1500 tweets at 95.75%, and Moderna vaccine with a dataset of 120 tweets at 96.88%. The analysis concludes that the public perception of these vaccines is generally positive.

关键词： COVID-19 Seminars Sentiment analysis Social networking (online) Blogs Vaccines Bayes methods

来源：评论

学校读者我要写书评

暂无评论

Live-Streaming-Based Dual-Teacher Classes for Equitable Education: Insights and Challenges from Local Teachers' Perspective in Disadvantaged Areas 25

Live-Streaming-Based Dual-Teacher Classes for Equitable Educ...

引用

2025 CHI Conference on Human Factors in Computing Systems, CHI 2025

作者： Sun, Yuling Chen, Jiaju Zhou, Xiaomu Ma, Xiaojuan Yao, Bingsheng Zhang, Kai He, Liang Wang, Dakuo Fudan University Shanghai China East China Normal University Shanghai China Informatics Program Northeastern University Boston MA United States Hong Kong University of Science and Technology Hong Kong Hong Kong Northeastern University Boston MA United States Shanghai Institution of AI for Education shanghai China School of Computer Science and Technology East China Normal University Shanghai China

ISBN: (纸本)9798400713941

Educational inequalities in disadvantaged areas have long been a global concern. While Information and Communication Technologies (ICTs) have shown great potential in addressing this issue, the unique challenges in disadvantaged areas often hinder the practical effectiveness of such technologies. This paper examines live-streaming-based dual-teacher classes (LSDC) through a qualitative study in disadvantaged regions of China. Our findings indicate that, although LSDC offers students in these regions access to high-quality educational resources, its practical implementation is fraught with challenges. Specifically, we foreground the pivotal role of local teachers in mitigating these challenges. Through a series of situated efforts, local teachers contextualize high-quality lectures to the local classroom environment, ensuring the expected educational outcomes. Based on our findings, we argue that greater recognition and support for the situational practices of local teachers is essential for fostering a more equitable, sustainable, and scalable technology-driven educational model in disadvantaged areas. © 2025 Copyright held by the owner/author(s). Publication rights licensed to ACM.

关键词： China Disadvantaged Area Education Education Equality Education Equity Live-streaming Remote Education Teachers Technology

来源：评论

学校读者我要写书评

暂无评论

A Comparative Study of Battery Energy Storage Systems in Isolated Grids: Novel Approach to Shave Peaks & Optimize Energy Costs

A Comparative Study of Battery Energy Storage Systems in Iso...

引用

International Conference on Power Engineering and Renewable Energy (ICPERE)

作者： Bagas Maulana Sutardi Ariesa Budi Zakaria Ilham Akbar Kevin Marojahan Banjar-Nahor Nanang Hariyanto Graduate Program PLN-ITB PLN Corporate University Jakarta Indonesia School of Electrical Engineering and Informatics Institut Teknologi Bandung (ITB) Bandung Indonesia

ISBN: (纸本)9781665455497

Isolated power systems are typically found in rural areas that are far separated from the power grid. Rapid decrease in BESS (Battery Energy Storage Systems) cost provides new opportunities and preferences for utilities to store cheap energy at low loads and power out that energy at peak loads. BESS is installed to shave or level demand during peak hours in an isolated system can avoid the use of peaking generations mostly coming from oil fuels. Considering the BESS function for peak shaving, the approach is to simulate the unit commitment of different battery sizes at 10 MW, because the use of oil fuels about 10 MW at peak load, so the simulation will be representing the total energy of the battery needed by the system. Based on the simulation, the life cycle cost is significantly greater than the benefit at all power and energy ratings. The purpose of this paper is to demonstrate the comprehensive technical and financial analysis needs to find the optimum battery type and capacity to provide optimum impact and financially benefit the system with an approach to shaving peaks and energy costs. A study-case method is demonstrated by taking an isolated system as the base and BESS performance can accelerate voltage in the subsystem for every substation when discharging at peak load. The simulation results that 10 MW/10 MWh lithium-ion batteries provide the highest benefit compared to the life cycle cost.

关键词： Renewable energy sources Costs Substations Oils Simulation Voltage Power grids

来源：评论

学校读者我要写书评

暂无评论

Classification of Paddy Leaf Disease Using MobileNet Model

Classification of Paddy Leaf Disease Using MobileNet Model

引用

International Conference on Computing, Engineering, and Design (ICCED)

作者： Fauzan Masykur Kusworo Adi Oky Dwi Nurhayati Doctoral Program Of Information System School Of Postgraduate Studies Diponegoro University Semarang Indonesia Department of Informatics Engineering Muhammadiyah University of Ponorogo Ponorogo Indonesia

ISBN: (纸本)9781665453905

Paddy is a rice-producing plant as a staple food raw material for most of the population in Indonesia. The rice planting process cannot be separated from pests and diseases that can cause crop failure and impact on rice production stocks. Currently, the process of monitoring the development of rice plants is done conventionally by direct observation by farmers. Meanwhile, in this paper, a classification model and determination of object detection in rice plants infected with pests and diseases is proposed based on leaf color. Leaf classification applies convolution conditional network algorithm with mobilenet as the architecture. The leaf image dataset was taken from public data from *** by involving 4 classes, namely healthy leaf images, brownspot, hispa and leafblast images. In each class the leaf images are grouped into 2 groups, namely 70% training and 30% validation. On the mobilenet backbone, it produces 97% accuracy and 22% loss with the time required for 195s 19s/step in 1 epoch round.

关键词： Training Image color analysis Sociology Computer architecture Production Object detection Raw materials

来源：评论

学校读者我要写书评

暂无评论

Modeling Non-Cooperative Dialogue: Theoretical and Empirical Insights

arXiv

引用

arXiv 2022年

作者： Sicilia, Anthony Maidment, Tristan Healy, Pat Alikhani, Malihe Intelligent Systems Program University of Pittsburgh Pittsburgh United States Department of Informatics and Networked Systems University of Pittsburgh Pittsburgh United States Department of Computer Science and Intelligent Systems Program University of Pittsburgh Pittsburgh United States

Investigating cooperativity of interlocutors is central in studying pragmatics of dialogue. Models of conversation that only assume cooperative agents fail to explain the dynamics of strategic conversations. Thus, we investigate the ability of agents to identify non-cooperative interlocutors while completing a concurrent visual-dialogue task. Within this novel setting, we study the optimality of communication strategies for achieving this multi-task objective. We use the tools of learning theory to develop a theoretical model for identifying non-cooperative interlocutors and apply this theory to analyze different communication strategies. We also introduce a corpus of non-cooperative conversations about images in the GuessWhat?! dataset proposed by De Vries et al. (2017). We use reinforcement learning to implement multiple communication strategies in this context and find empirical results validate our theory. © 2022, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Human Learners Integrate Visual and Linguistic Information in Cross-Situational Verb Learning 43

Human Learners Integrate Visual and Linguistic Information i...

引用

43rd Annual Meeting of the Cognitive Science Society: Comparative Cognition: Animal Minds, CogSci 2021

作者： Zhang, Yayun Amatuni, Andrei Cain, Ellis Wang, Xizi Crandall, David Yu, Chen Department of Psychology University of Texas Austin United States Cognitive Science Program Indiana University Bloomington United States Luddy School of Informatics Computing and Engineering Indiana University Bloomington United States

Learning verbs is challenging because it is difficult to infer the precise meaning of a verb when there are a multitude of relations that one can derive from a single event. To study this verb learning challenge, we used children’s egocentric view collected from naturalistic toy-play interaction as learning materials and investigated how visual and linguistic information provided in individual naming moments as well as cross-situational information provided from multiple learning moments can help learners resolve this mapping problem using the Human Simulation Paradigm. Our results show that learners benefit from seeing children’s egocentric views compared to third-person observations. In addition, linguistic information can help learners identify the correct verb meaning by eliminating possible meanings that do not belong to the linguistic category. Learners are also able to integrate visual and linguistic information both within and across learning situations to reduce the ambiguity in the space of possible verb meanings. © Cognitive Science Society: Comparative Cognition: Animal Minds, CogSci *** rights reserved.

关键词： Toys

来源：评论

学校读者我要写书评

暂无评论

Hamiltonian Cycles for Finite Weyl Groupoids

arXiv

引用

arXiv 2023年

作者： Inoue, Takato Yamane, Hiroyuki Mathematics and Informatics Program Graduate School of Science and Engineering University of Toyama 3190 Gofuku Toyama930-8555 Japan Faculty of Science Academic Assembly University of Toyama 3190 Gofuku Toyama930-8555 Japan

Let Γ(W) be the Cayley graph of a finite Weyl groupoid W. In this paper, we show an existence of a Hamitonian cycle of Γ(W) for any W. We exatctly draw a Hamiltonian cycle of Γ(W) for any (resp. some) irreducible W of rank three (resp. four). Moreover for the irreducible W of rank three, we give a second largest eigenvalue of the adjacency matrix of Γ(W), and know if Γ(W) is a bipartite Ramanujan graph or not. Copyright © 2023, The Authors. All rights reserved.

关键词： Hamiltonians

来源：评论

学校读者我要写书评

暂无评论

CORNN: convex optimization of recurrent neural networks for rapid inference of neural dynamics 23

CORNN: convex optimization of recurrent neural networks for ...

引用

Proceedings of the 37th International Conference on Neural Information Processing Systems

作者： Fatih Dinc Adam Shai Mark J. Schnitzer Hidenori Tanaka Department of Applied Physics Stanford University Stanford CA CNC Program Stanford University Stanford CA Howard Hughes Medical Institute CNC Program Stanford University Stanford CA Physics & Informatics Laboratories NTT Research Inc. Sunnyvale CA and Center for Brain Science Harvard University Cambridge MA

Advances in optical and electrophysiological recording technologies have made it possible to record the dynamics of thousands of neurons, opening up new possibilities for interpreting and controlling large neural populations in behaving animals. A promising way to extract computational principles from these large datasets is to train data-constrained recurrent neural networks (dRNNs). Performing this training in real-time could open doors for research techniques and medical applications to model and control interventions at single-cell resolution and drive desired forms of animal behavior. However, existing training algorithms for dRNNs are inefficient and have limited scalability, making it a challenge to analyze large neural recordings even in offline scenarios. To address these issues, we introduce a training method termed Convex Optimization of Recurrent Neural Networks (CORNN). In studies of simulated recordings, CORNN attained training speeds ~ 100-fold faster than traditional optimization approaches while maintaining or enhancing modeling accuracy. We further validated CORNN on simulations with thousands of cells that performed simple computations such as those of a 3-bit flip-flop or the execution of a timed response. Finally, we showed that CORNN can robustly reproduce network dynamics and underlying attractor structures despite mismatches between generator and inference models, severe subsampling of observed neurons, or mismatches in neural time-scales. Overall, by training dRNNs with millions of parameters in subminute processing times on a standard computer, CORNN constitutes a first step towards real-time network reproduction constrained on large-scale neural recordings and a powerful computational tool for advancing the understanding of neural computation. CORNN software and reproduction code are available at https://***/schnitzer-lab/CORNN-public

关键词：

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：