检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Steyvers, Mark Tejeda, Heliodoro Kumar, Aakriti Belem, Catarina Karny, Sheer Hu, Xinyue Mayer, Lukas Smyth, Padhraic Department of Cognitive Sciences University of California Irvine United States Department of Computer Science University of California Irvine United States

As artificial intelligence systems, particularly large language models (LLMs), become increasingly integrated into decision-making processes, the ability to trust their outputs is crucial. To earn human trust, LLMs must be well calibrated such that they can accurately assess and communicate the likelihood of their predictions being correct. Whereas recent work has focused on LLMs’ internal confidence, less is understood about how effectively they convey uncertainty to users. Here we explore the calibration gap, which refers to the difference between human confidence in LLM-generated answers and the models’ actual confidence, and the discrimination gap, which reflects how well humans and models can distinguish between correct and incorrect answers. Our experiments with multiple-choice and short-answer questions reveal that users tend to overestimate the accuracy of LLM responses when provided with default explanations. Moreover, longer explanations increased user confidence, even when the extra length did not improve answer accuracy. By adjusting LLM explanations to better reflect the models’ internal confidence, both the calibration gap and the discrimination gap narrowed, significantly improving user perception of LLM accuracy. These findings underscore the importance of accurate uncertainty communication and highlight the effect of explanation length in influencing user trust in artificial-intelligence-assisted decision-making environments. © 2024, CC BY-SA.

关键词： Calibration

来源：评论

学校读者我要写书评

暂无评论

Exciting Action: Investigating Efficient Exploration for Learning Musculoskeletal Humanoid Locomotion

arXiv

引用

arXiv 2024年

作者： Geiß, Henri-Jacques Al-Hafez, Firas Seyfarth, Andre Peters, Jan Tateo, Davide Computer Science Department Technical University Darmstadt Germany Human Sciences Department Technical University Darmstadt Germany Centre for Cognitive Science Hessian.AI Germany

Learning a locomotion controller for a musculoskeletal system is challenging due to over-actuation and high-dimensional action space. While many reinforcement learning methods attempt to address this issue, they often struggle to learn human-like gaits because of the complexity involved in engineering an effective reward function. In this paper, we demonstrate that adversarial imitation learning can address this issue by analyzing key problems and providing solutions using both current literature and novel techniques. We validate our methodology by learning walking and running gaits on a simulated humanoid model with 16 degrees of freedom and 92 Muscle-Tendon Units, achieving natural-looking gaits with only a few demonstrations. Code is available at https://***/henriTUD/musculoco learning. © 2024, CC BY-SA.

关键词： Musculoskeletal system

来源：评论

学校读者我要写书评

暂无评论

Weather Prediction in Agriculture Yields with Transformer Model 9th

Weather Prediction in Agriculture Yields with Transformer Mo...

引用

9th International Conference on computer science and Computational Intelligence, ICCSCI 2024

作者： Arifin, Yulyani Sonata, Ilvico Maryani Gunawan, Elizabeth Paskahlia Computer Science Department Binus Graduate Program Bina Nusantara University Jl. K.H. Syahdan No.9 Kemanggisan Palmerah Jakarta11480 Indonesia Information System Department School of Information System Bina Nusantara University Jl. K.H. Syahdan No.9 Kemanggisan Palmerah Jakarta11480 Indonesia Computer Science Department School of Computer Science Bina Nusantara University Jl. K.H. Syahdan No.9 Kemanggisan Palmerah Jakarta11480 Indonesia

Agriculture has a lot of relations with SDG from United Nations especially in end hungers and sustainable agriculture. One of factor important in agriculture is weather. Weather prediction is very important in agriculture. Weather greatly influences agricultural yields and the types of crops to be planted. Weather predictions, especially temperature and humidity, can be done using time series data analysis. Current developments in deep learning enable more accurate predictions of temperature and humidity based on time series dataset. Some deep learning models can be used to predict data based on training using time series data. The deep programming models used include Gated Recurrent Unit (GRU), Recurrent Neural Network (RNN), and Long Short-Term Memory (LSTM). The Transformer deep learning model, which was first introduced in 2017, is a sequence-to-sequence model which is very suitable for predicting time series data. This paper will explain temperature and humidity prediction using the Transformer deep learning model. Experimental results show that the Transformer model can predict temperature and humidity based on training results using a time series dataset. The accuracy obtained by the Transformer deep learning model is better compared to the LSTM deep learning model as indicated by lower MAE, MSE and RMSE values compared to the LSTM model. © 2024 The Authors.

关键词： Prediction models

来源：评论

学校读者我要写书评

暂无评论

Hidden or Inferred: Fair Learning-to-Rank with Unknown Demographics 24

Hidden or Inferred: Fair Learning-to-Rank with Unknown Demog...

引用

Proceedings of the 2024 AAAI/ACM Conference on AI, Ethics, and Society

作者： Oluseun Olulana Kathleen Cachel Fabricio Murai Elke Rundensteiner Data Science Program Worcester Polytechnic Institute Worcester MA Data Science Program Worcester Polytechnic Institute Worcester MA and Computer Science Department Worcester Polytechnic Institute Worcester MA

As learning-to-rank models are increasingly deployed for decision-making in areas with profound life implications, the FairML community has been developing fair learning-to-rank (LTR) models. These models rely on the availability of sensitive demographic features such as race or sex. However, in practice, regulatory obstacles and privacy concerns protect this data from collection and use. As a result, practitioners may either need to promote fairness despite the absence of these features or turn to demographic inference tools to attempt to infer them. Given that these tools are fallible, this paper aims to further understand how errors in demographic inference impact the fairness performance of popular fair LTR strategies. In which cases would it be better to keep such demographic attributes hidden from models versus infer them? We examine a spectrum of fair LTR strategies ranging from fair LTR with and without demographic features hidden versus inferred to fairness-unaware LTR followed by fair re-ranking. We conduct a controlled empirical investigation modeling different levels of inference errors by systematically perturbing the inferred sensitive attribute. We also perform three case studies with real-world datasets and popular open-source inference methods. Our findings reveal that as inference noise grows, LTR-based methods that incorporate fairness considerations into the learning process may increase bias. In contrast, fair re-ranking strategies are more robust to inference errors. All source code, data, and experimental artifacts of our experimental study are available here: https://***/sewen007/***

关键词：

来源：评论

学校读者我要写书评

暂无评论

Predicting Confidentiality, Integrity, and Availability from SQL Injection Payload 7

Predicting Confidentiality, Integrity, and Availability from...

引用

7th International Conference on Information Management and Technology, ICIMTech 2022

作者： Muliono, Yohan Darus, Mohamad Yusof Pardomuan, Chrisando Ryan Ariffin, Muhammad Azizi Mohd Kurniawan, Aditya School of Computer Science Bina Nusantara University Cyber Security Program Computer Science Department Jakarta11480 Indonesia Universiti Teknologi Mara Faculty of Computer and Mathematical Sciences Shah Alam40450 Malaysia

ISBN: (数字)9781665450904

ISBN: (纸本)9781665450904

SQL Injection has been around as a harmful and prolific threat on web applications for more than 20 years, yet it still poses a huge threat to the World Wide Web. Rapidly evolving web technology has not eradicated this threat;In 2017 51 % of web application attacks are SQL injection attacks. Most conventional practices to prevent SQL injection attacks revolves around secure web and database programming and administration techniques. Despite developer ignorance, a large number of online applications remain susceptible to SQL injection attacks. There is a need for a more effective method to detect and prevent SQL Injection attacks. In this research, we offer a unique machine learning-based strategy for identifying potential SQL injection attack (SQL injection attack) threats. Application of the proposed method in a Security Information and Event Management(SIEM) system will be discussed. SIEM can aggregate and normalize event information from multiple sources, and detect malicious events from analysis of these information. The result of this work shows that a machine learning based SQL injection attack detector which uses SIEM approach possess high accuracy in detecting malicious SQL queries. © 2022 IEEE.

关键词： Support vector machines

来源：评论

学校读者我要写书评

暂无评论

Enhancing cybersecurity resilience through advanced red-teaming exercises and MITRE ATT&CK framework integration: A paradigm shift in cybersecurity assessment

Cyber Security and Applications

引用

Cyber Security and Applications 2025年 3卷

作者： Yulianto, Semi Soewito, Benfano Gaol, Ford Lumban Kurniawan, Aditya Computer Science Department Binus Graduate Program Doctor of Computer Science Bina Nusantara University Jakarta 11480 Indonesia

As cybersecurity threats evolve alarmingly, conventional defense strategies are becoming increasingly ineffective. In response to this urgent challenge, our study presents a transformative approach to red-teaming exercises by integrating the MITRE ATT&CK framework. This innovative integration leverages real-world attacker tactics and behaviors to create highly realistic scenarios that rigorously test defenses and uncover previously unidentified vulnerabilities. Our comprehensive evaluation demonstrates a significant enhancement in the realism and effectiveness of red-teaming, leading to improved vulnerability identification and the generation of actionable insights for proactive remediation. This study uniquely contributes by providing a structured, data-driven methodology that aligns current defenses with emerging threat tactics and promotes continuous improvement in cybersecurity resilience. Such advancements are essential for organizations seeking to stay ahead of attackers and maintain robust cyber defenses in dynamic threat environments. The significance of this study lies in offering a structured approach to strengthen cybersecurity resilience against evolving threats and providing a model for continual defense enhancement. Future research should investigate the impact of this integration on long-term cybersecurity strategies. © 2024 The Authors

关键词： Continuous defense improvement Cyber resilience Cybersecurity MITRE ATT&CK framework Red-teaming exercises

来源：评论

学校读者我要写书评

暂无评论

The "LLM World of Words" English free association norms generated by large language models

arXiv

引用

arXiv 2024年

作者： Abramski, Katherine Improta, Riccardo Rossetti, Giulio Stella, Massimo University of Pisa Department of Computer Science Pisa Italy University of Trento Department of Psychology and Cognitive Science Trento Italy National Research Council of Italy Institute of Information Science and Technologies Pisa Italy

Free associations have been extensively used in cognitive psychology and linguistics for studying how conceptual knowledge is organized. Recently, the potential of applying a similar approach for investigating the knowledge encoded in LLMs has emerged, specifically as a method for investigating LLM biases. However, the absence of large-scale LLM-generated free association norms that are comparable with human-generated norms is an obstacle to this new research direction. To address this limitation, we create a new dataset of LLM-generated free association norms modeled after the "Small World of Words" (SWOW) human-generated norms consisting of approximately 12,000 cue words. We prompt three LLMs, namely Mistral, Llama3, and Haiku, with the same cues as those in the SWOW norms to generate three novel comparable datasets, the "LLM World of Words" (LWOW). Using both SWOW and LWOW norms, we construct cognitive network models of semantic memory that represent the conceptual knowledge possessed by humans and LLMs. We demonstrate how these datasets can be used for investigating implicit biases in humans and LLMs, such as the harmful gender stereotypes that are prevalent both in society and LLM outputs. © 2024, CC BY.

关键词： Semantics

来源：评论

学校读者我要写书评

暂无评论

Investigating the User Experience and Identifying User Needs for Kitchen Appliances Using Thematic Analysis

Investigating the User Experience and Identifying User Needs...

引用

2022 IEEE International Conference on Industrial Engineering and Engineering Management, IEEM 2022

作者： Jung, W.C. Lazaro, M.J. Lee, J.H. Kim, G.B. Yang, E.J. Yun, M.H. Seoul National University Department of Industrial Engineering Seoul Korea Republic of Seoul National University Interdisciplianary Program in Cognitive Science Seoul Korea Republic of

ISBN: (数字)9781665486873

ISBN: (纸本)9781665486873

Kitchen appliances are essential to accomplish cooking tasks efficiently. Advancements in technology have led to changes in the needs and expectations of the users of commonly used kitchen appliances. Thus, this qualitative research aims to investigate the user experience and identify the user needs for common kitchen appliances (microwave, cooktop, oven, and dishwasher) by conducting semi-structured interviews with users who frequently use the appliances. Results of the thematic analysis showed that UX factors such as informativeness, comfortability, and safety were prominent sub-themes across all kitchen appliances. Based on the findings, we presented design recommendations on how the product can be improved. © 2022 IEEE.

关键词： Kitchens

来源：评论

学校读者我要写书评

暂无评论

Trends in Research Publication Topics Related to Artificial Intelligence for Medicine in Medical Education: A Bibliometric Analysis 25

Trends in Research Publication Topics Related to Artificial ...

引用

25th International Seminar on Intelligent Technology and Its Applications, ISITIA 2024

作者： Maulana, Fairuz Iqbal Lestari, Dian Adi, Puput Dani Prasetyo Purnomo, Agung Amalia, Mufidah Nur Hamim, Miftahul Bina Nusantara University School of Computer Science Computer Science Department Jakarta11480 Indonesia Airlangga University Faculty of Medicine Surabaya60132 Indonesia Research Center for Telecommunication Bandung Indonesia Bina Nusantara University Binus Business School Undergraduate Program Entrepreneurship Department Jakarta11480 Indonesia Bina Nusantara University Faculty of Humanities Digital Language Learning Center Computer Science Department Jakarta11480 Indonesia Bina Nusantara University Entrepreneurship Business Creation Binus Business School Jakarta11480 Indonesia

ISBN: (纸本)9798350378573

This bibliometric analysis aims to explore the trends in research publication topics related to Artificial Intelligence (AI) for medicine in medical education over the last five years (2019-2023). By synthesizing data from a variety of scholarly articles, this study seeks to identify key themes, emerging areas of interest, and research trends within this specific domain. The research question focuses on understanding the evolution of research topics in AI for medicine in medical education, while the objective is to conduct a comprehensive analysis using bibliometric techniques. Data selection from the Scopus database yielded a total of 389 documents from 290 sources. There are 1889 Authors, with author collaboration points on Co-Authors per Doc of 5.11. Regarding the quantity of published documents, the USA, China, and Canada have the highest documents. The software employed for bibliometric analysis were R and VOSviewer. By leveraging bibliometric analysis, this research aims to address the existing gap in the literature and provide valuable insights for researchers, educators, and practitioners interested in AI applications in medical education. The present solution exploration will offer recommendations for future research directions and priorities within the field, contributing to the advancement of knowledge and practice in the integration of AI technologies in medical education. © 2024 IEEE.

关键词： Bibliographies

来源：评论

学校读者我要写书评

暂无评论

Ensuring Ethical, Transparent, and Auditable Use of Education Data and Algorithms on AutoML 23

Ensuring Ethical, Transparent, and Auditable Use of Educatio...

引用

Proceedings of the 2023 6th International Conference on Machine Learning and Natural Language Processing

作者： Kylie Griep Leah Stone Stephen Edwards Malik Hill Jialin Cui Chengyuan Liu Qinjin Jia Yang Song Department of Computer Science University of North Carolina Wilmington USA Master of Data Science Program University of North Carolina Wilmington USA Department of Computer Science NC State University USA

ISBN: (纸本)9798400709241

Automated machine learning (AutoML) creates additional opportunities for less advanced users to build and test their own data mining models. Even though AutoML creates the models for the user, there is still technical knowledge and tools needed to evaluate those models, and due to the black-box nature of the machine learning models, problems can arise with regard to algorithmic biases and fairness. Such biases can escalate in future applications, necessitating a structured approach for fairness evaluation in AutoML. This involves defining fairness criteria, selecting appropriate metrics, assessing fairness across groups, and addressing biases. In the realm of educational data mining, where AutoML is prevalent, biases related to attributes like gender or race can lead to unethical outcomes. Since fairness metrics vary in definition and strength, and some may even contradict others, making fairness evaluation more complex. In this paper, ten fairness metrics were chosen, explored, and implemented on four AutoML tools, Vertex AI, AutoSklearn, AutoKeras, and PyCaret. We identified two open educational datasets and built both prediction and classification models on those AutoML frameworks. We report our work in evaluating different machine learning models created by AutoML and provide discussions about the challenges in evaluating fairness in those models and our effort to mitigate and resolve the problems of algorithmic bias in educational data mining.

关键词： AutoML Automated machine learning algorithmic bias educational data mining fairness

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：