检索结果-内蒙古大学图书馆

International Conference on Computing and Networking Technology (ICCNT)

作者： R. Meena B. Gunasundari J. Omana N. Sripriya G. Yamini Kp Revathi Department of AI&DS Easwari Engineering College Chennai India Department of CSE Prathyusha Engineering College Chennai India School of Computer Science and Engineering VIT University Chennai India Department of English Veltech Rangarajan Dr.Sagunthala R&D Institute of Science and Technology Chennai India

ISBN: (数字)9798350370249

ISBN: (纸本)9798350370270

This study aims to identify critical predictors of suicide and inform targeted interventions using data from 27,014 college students. It examines a comprehensive range of mental health indicators, including psychiatric symptoms, suicidal intentions, impulsivity, sleep disturbances, internet addiction, aggression, self-injury behaviors, and eating problems, alongside demographic and socioeconomic variables. Analytical techniques, including exploratory data analysis, cross-tabulation, and correlation analysis, uncover the complex interplay between these factors and suicide risk. The study evaluates several machine learning algorithms. The weighted Random Forest uses feature weighting based on point-biserial correlation coefficients to enhance prediction accuracy and prioritize relevant features. The findings reveal significant associations between higher scores in mental health indicators and increased suicidal behavior, emphasizing the need for comprehensive mental health support systems. By identifying key risk factors through machine learning models, the study provides actionable insights for developing targeted interventions aimed at reducing suicide risk among college students.

关键词： Measurement Accuracy Correlation Machine learning algorithms Mental health Robustness Risk management Resource management Socioeconomics Random forests

来源：评论

学校读者我要写书评

暂无评论

TOWARDS UNIVERSAL SPEECH DISCRETE TOKENS: A CASE STUDY FOR ASR AND TTS

arXiv

引用

arXiv 2023年

作者： Yang, Yifan Shen, Feiyu Du, Chenpeng Ma, Ziyang Yu, Kai Povey, Daniel Chen, Xie MoE Key Lab of Artificial Intelligence AI Institute X-LANCE Lab Department of Computer Science and Engineering Shanghai Jiao Tong University China Xiaomi Corporation Beijing China

Self-supervised learning (SSL) proficiency in speech-related tasks has driven research into utilizing discrete tokens for speech tasks like recognition and translation, which offer lower storage requirements and great potential to employ natural language processing techniques. However, these studies, mainly single-task focused, faced challenges like overfitting and performance degradation in speech recognition tasks, often at the cost of sacrificing performance in multi-task scenarios. This study presents a comprehensive comparison and optimization of discrete tokens generated by various leading SSL models in speech recognition and synthesis tasks. We aim to explore the universality of speech discrete tokens across multiple speech tasks. Experimental results demonstrate that discrete tokens achieve comparable results against systems trained on FBank features in speech recognition tasks and outperform mel-spectrogram features in speech synthesis in subjective and objective metrics. These findings suggest that universal discrete tokens have enormous potential in various speech-related tasks. Our work is open-source and publicly available at https://***/k2-fsa/icefall. Index Terms- self-supervised learning, discrete tokens, speech recognition, text-to-speech. Copyright © 2023, The Authors. All rights reserved.

关键词： Speech recognition

来源：评论

学校读者我要写书评

暂无评论

Have Large Language Models Developed a Personality?: Applicability of Self-Assessment Tests in Measuring Personality in LLMs

arXiv

引用

arXiv 2023年

作者： Song, Xiaoyang Gupta, Akshat Mohebbizadeh, Kiyan Hu, Shujie Singh, Anant Data Science Institute Columbia University United States J.P. Morgan AI Research United States Department of Computer Engineering New York University United States

Have Large Language Models (LLMs) developed a personality? The short answer is a resounding "We Don’t Know!". In this paper, we show that we do not yet have the right tools to measure personality in language models. Personality is an important characteristic that influences behavior. As LLMs emulate human-like intelligence and performance in various tasks, a natural question to ask is whether these models have developed a personality. Previous works have evaluated machine personality through self-assessment personality tests, which are a set of multiple-choice questions created to evaluate personality in humans. A fundamental assumption here is that human personality tests can accurately measure personality in machines. In this paper, we investigate the emergence of personality in five LLMs of difference sizes ranging from 1.5B to 30B. We propose the Option-Order Symmetry property as a necessary condition for reliability of these self-assessment tests. Under this condition, the answer to self-assessment questions are invariant to the order in which the options are presented. We find that many LLMs personality test responses do not preserve option-order symmetry. We then take a deeper look at LLMs test responses where option-order symmetry is preserved to find that in these cases, LLMs do not take into account the situational statement being tested and produce the exact same answer irrespective of the situation being tested. We also identify the existence of inherent biases in these language models which is the root cause of the aforementioned phenomenon and makes self-assessment tests unreliable. These observations indicate that self-assessment tests are not the correct tools to measure personality in language models. Through this paper, we hope to draw attention to the shortcomings of current literature in measuring personality in LLMs and call for developing specific tools for machine personality measurement. Copyright © 2023, The Authors. All rights reserved.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Augmenting Data Analysis Through Innovative Visualization Techniques

Augmenting Data Analysis Through Innovative Visualization Te...

引用

International Conference on Futuristic Technologies (INCOFT)

作者： Uma Rani K Omkaresh S. Kulkarn Ravi Kant Pareek Trapty Agrawal R Murugan N.V. Balaji Department of Information Technology Prince Shri Venkateshwara Padmavathy Engineering College Chennai Department of Civil Engineering Vivekananda Global University Jaipur India Department of Computer Science and Information Technology Jain (Deemed to be University) Bangalore Karnataka India Department of Computer Science and Engineering (AI & ML) Vishwakarma Institute of Information Technology Pune India Maharishi School of Engineering and Technology Maharishi University of Information Technology Uttar Pradesh India Department of Computer Science Karpagam Academy of Higher Education Coimbatore

This paper examines new and progressive methods in information visualization for analysis and selectionmaking. It explains various traits of records that might be made less burdensome to apprehend via visuals, which include graphs, pictures, and other visible representations. It highlights the strength of facts visualization to assist people in interpreting, advantage insights, and conclude from information, making them more green and powerful in their evaluation. The paper covers the significance of records visualization modern and growing techniques and examines how facts analysis and choice-making may be augmented using these stylish visuals. Researchers can use statistics visualization techniques to reach broader audiences and help them understand important information more correctly.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Planing It by Ear: Convolutional Neural Networks for Acoustic Anomaly Detection in Industrial Wood Planers

Planing It by Ear: Convolutional Neural Networks for Acousti...

引用

2025 IEEE International Conference on Acoustics, Speech, and Signal Processing, ICASSP 2025

作者： Deschênes, Anthony Georges, Rémi Subakan, Cem Ugulino, Bruna Henry, Antoine Morin, Michael Dep. of Wood and Forest Sciences Université Laval QC Canada Dep. of Computer Science and Software Engineering Université Laval QC Canada Dep. of Operations and Decision Systems Université Laval QC Canada FORAC Research Consortium Université Laval QC Canada Mila-Quebec AI Institute QC Canada FPInnovations QC Canada Concordia University QC Canada

ISBN: (纸本)9798350368741

In recent years, the wood product industry has been facing a skilled labor shortage. The result is more frequent sudden failures, resulting in additional costs for these companies already operating in a very competitive market. Moreover, sawmills are challenging environments for machinery and sensors. Given that experienced machine operators may be able to diagnose defects or malfunctions, one possible way of assisting novice operators is through acoustic monitoring. As a step towards the automation of wood-processing equipment and decision support systems for machine operators, in this paper, we explore using a deep convolutional autoencoder for acoustic anomaly detection of wood planers on a new real-life dataset. Specifically, our convolutional autoencoder with skip connections (Skip-CAE) and our Skip-CAE transformer outperform the DCASE autoencoder baseline, one-class SVM, isolation forest and a published convolutional autoencoder architecture, respectively obtaining an area under the ROC curve of 0.846 and 0.875 on a dataset of real-factory planer sounds. Moreover, we show that adding skip connections and attention mechanism under the form of a transformer encoder-decoder helps to further improve the anomaly detection capabilities. © 2025 IEEE.

关键词： Acoustic Anomaly detection Convolutional neural networks Transformer Wood planer

来源：评论

学校读者我要写书评

暂无评论

A Hybrid CNN-GRU model for Session-based Recommender Systems

A Hybrid CNN-GRU model for Session-based Recommender Systems

引用

International Conference on computer and Communications (ICCC)

作者： Abebe Tamrat Tegene Qiao Liu Yanglei Gan Elias Jimale Melak Ayenew Habte Leka School of Computer Science and Engineering University of Electronic Science and Technology of China Chengdu China Center of Excellence of AI and Robotics Addis Ababa Science and Technology University Addis Ababa Ethiopia School of Information and Software Engineering University of Electronic Science and Technology of China Chengdu China

In many e-commerce recommendation systems, user profiles are mostly not available. In such cases, the recommendation task considers user behaviors like click sessions or browsing history. Traditional recommender methods often rely on users’ past interactions for recommending user interests, which is difficult to apply to session-based recommender systems. However, these methods perform very poorly on this task. Recently, deep learning techniques such as recurrent neural networks (RNNs) and convolutional neural networks (CNNs) showed outstanding performance in session-based recommendation problems. In this paper, we propose a hybrid CNN and gated recurrent unit (GRU) model that exploits the advantages of both CNN and GRU in modeling the rigid order requirement for interactions inside sessions and modeling the sequential long-term dependencies in session data, respectively. Our proposed hybrid model has two architectures: the first part consists of the CNN layers that take input from Item ID to learn local features and relationships in sessions, and the second part is a GRU layer that takes the output from the CNN layer to learn the long-term dependencies in that session. This hybrid technique enables the architecture for modeling prominent features in the session to provide accurate recommendations. Extensive experiments over a real-world dataset demonstrate that the proposed hybrid model performs well and outperforms those GRU and CNN-based models in the state-of-the-art recommendation task on both metrics used.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ROBUST EARLY-LEARNING: HINDERING THE MEMORIZATION OF NOISY LABELS 9

ROBUST EARLY-LEARNING: HINDERING THE MEMORIZATION OF NOISY L...

引用

9th International Conference on Learning Representations, ICLR 2021

作者： Xia, Xiaobo Liu, Tongliang Han, Bo Gong, Chen Wang, Nannan Ge, Zongyuan Chang, Yi Trustworthy Machine Learning Lab School of Computer Science The University of Sydney Australia Department of Computer Science Hong Kong Baptist University Hong Kong School of Computer Science and Engineering Nanjing University of Science and Technology China ISN State Key Laboratory School of Telecommunications Engineering Xidian University China Medical AI Group Faculty of Engineering Monash University Australia Airdoc Research Monash University Australia School of Artificial Intelligence Jilin University China

The memorization effects of deep networks show that they will first memorize training data with clean labels and then those with noisy labels. The early stopping method therefore can be exploited for learning with noisy labels. However, the side effect brought by noisy labels will influence the memorization of clean labels before early stopping. In this paper, motivated by the lottery ticket hypothesis which shows that only partial parameters are important for generalization, we find that only partial parameters are important for fitting clean labels and generalize well, which we term as critical parameters;while the other parameters tend to fit noisy labels and cannot generalize well, which we term as non-critical parameters. Based on this, we propose robust early-learning to reduce the side effect of noisy labels before early stopping and thus enhance the memorization of clean labels. Specifically, in each iteration, we divide all parameters into the critical and non-critical ones, and then perform different update rules for different types of parameters. Extensive experiments on benchmark-simulated and real-world label-noise datasets demonstrate the superiority of the proposed method over the state-of-the-art label-noise learning methods. © 2021 ICLR 2021 - 9th International Conference on Learning Representations. All rights reserved.

关键词： Iterative methods

来源：评论

学校读者我要写书评

暂无评论

airLine: Efficient Learnable Line Detection with Local Edge Voting

AirLine: Efficient Learnable Line Detection with Local Edge ...

引用

IEEE/RSJ International Conference on Intelligent Robots and Systems (IROS)

作者： Xiao Lin Chen Wang College of Computing Georgia Institute of Technology Atlanta GA USA Department of Computer Science and Engineering Spatial AI & Robotics Lab State University of New York at Buffalo NY USA

Line detection is widely used in many robotic tasks such as scene recognition, 3D reconstruction, and simultaneous localization and mapping (SLAM). Compared to points, lines can provide both low-level and high-level geometrical information for downstream tasks. In this paper, we propose a novel learnable edge-based line detection algorithm, airLine, which can be applied to various tasks. In contrast to existing learnable endpoint-based methods, which are sensitive to the geometrical condition of environments, airLine can extract line segments directly from edges, resulting in a better generalization ability for unseen environments. To balance efficiency and accuracy, we introduce a region-grow algorithm and a local edge voting scheme for line parameterization. To the best of our knowledge, airLine is one of the first learnable edge-based line detection methods. Our extensive experiments have shown that it retains state-of-the-art-Ievel precision, yet with a $3-80\times$ runtime acceleration compared to other learning-based methods, which is critical for low-power robots.

关键词：

来源：评论

学校读者我要写书评

暂无评论

DIVERSITY-AWARE REINFORCEMENT LEARNING FOR de novo DRUG DESIGN

arXiv

引用

arXiv 2024年

作者： Svensson, Hampus Gummesson Tyrchan, Christian Engkvist, Ola Chehreghani, Morteza Haghir Department of Computer Science and Engineering Chalmers University of Technology University of Gothenburg Gothenburg Sweden Molecular AI Discovery Sciences R&D AstraZeneca Gothenburg Sweden BioPharmaceuticals R&D AstraZeneca Gothenburg Sweden

Fine-tuning a pre-trained generative model has demonstrated good performance in generating promising drug molecules. The fine-tuning task is often formulated as a reinforcement learning problem, where previous methods efficiently learn to optimize a reward function to generate potential drug molecules. Nevertheless, in the absence of an adaptive update mechanism for the reward function, the optimization process can become stuck in local optima. The efficacy of the optimal molecule in a local optimization may not translate to usefulness in the subsequent drug optimization process or as a potential standalone clinical candidate. Therefore, it is important to generate a diverse set of promising molecules. Prior work has modified the reward function by penalizing structurally similar molecules, primarily focusing on finding molecules with higher rewards. To date, no study has comprehensively examined how different adaptive update mechanisms for the reward function influence the diversity of generated molecules. In this work, we investigate a wide range of intrinsic motivation methods and strategies to penalize the extrinsic reward, and how they affect the diversity of the set of generated molecules. Our experiments reveal that combining structure- and prediction-based methods generally yields better results in terms of molecular diversity. © 2024, CC BY.

关键词： Drug discovery

来源：评论

学校读者我要写书评

暂无评论

Intrusion Detection System in Wireless Sensor Networks using Modified Recurrent Neural Network with Long Short-Term Memory

Intrusion Detection System in Wireless Sensor Networks using...

引用

Integrated Circuits and Communication Systems (ICICACS), IEEE International Conference on

作者： K. Ramkumar Laith H. Alzubaidi V Malathy T. Venkatesh Kruthika C G Department of Computer Science and Engineering SRM Institute of Science and Technology Chennai India The Islamic University Najaf Iraq Department of ECE School of Engineering SR University Telangana India Department of ECE KPR Institute of Engineering and Technology Coimbatore India Department of AI and ML Nitte Meenakshi Institute of Technology Bengaluru India

Leveraging deep learning, Intrusion Detection Systems (IDS) for Wireless Sensor Networks (WSN) enhance security through advanced analytics, detecting and mitigating network intrusions in real-time for robust protection. Evaluations and scalability considerations are critical for improving the applicability and robustness of intrusion detection systems based on deep learning techniques. This paper introduces a novel approach: a modified Recurrent Neural Network with Long Short-Term Memory (RNN-LSTM) for IDS in WSN classification, utilizing the NSL-KDD dataset. Rigorous performance validation demonstrates outstanding metrics, including an accuracy of 99.95%, precision at 99.93%, and a robust recall of 95.69%, with the F1-score reaching 99.80%, surpassing benchmarks set by existing methods. Notably, current approaches like Deep Neural Network (DNN), Convolutional Neural Network-LSTM (CNN-LSTM), Conditional Generative Adversarial Network (CGAN), and Stacked Autoencoders-DNN (SAE-DNN) face limitations in achieving optimal accuracy and robustness. The proposed RNN-LSTM method exhibits significant advancements in addressing these challenges for effective intrusion detection in WSN

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：