检索结果-内蒙古大学图书馆

CleanRL: High-quality Single-file Implementations of Deep Reinforcement Learning Algorithms

Journal of Machine Learning Research 2022年第1期23卷 1-18页

作者： Huang, Shengyi Julien Dossa, Rousslan Fernand Ye, Chang Braga, Jeff Chakraborty, Dipam Mehta, Kinal Araújo, João G.M. College of Computing and Informatics Drexel University United States Graduate School of System Informatics Kobe University Japan Tandon School of Engineering New York University United States AIcrowd International Institute of Information Technology Hyderabad India Cohere

CleanRL is an open-source library that provides high-quality single-file implementations of Deep Reinforcement Learning (DRL) algorithms. These single-file implementations are self-contained algorithm variant files such as ***, ***, and ppo *** that individually include all algorithm variant’s implementation details. Such a paradigm significantly reduces the complexity and the lines of code (LOC) in each implemented variant, which makes them quicker and easier to understand. This paradigm gives the researchers the most fine-grained control over all aspects of the algorithm in a single file, allowing them to prototype novel features quickly. Despite having succinct implementations, CleanRL’s codebase is thoroughly documented and benchmarked to ensure performance is on par with reputable sources. As a result, CleanRL produces a repository tailor-fit for two purposes: 1) understanding all implementation details of DRL algorithms and 2) quickly prototyping novel features. CleanRL’s source code can be found at https://***/vwxyzjn/cleanrl. ©2022 Shengyi Huang, Rousslan Fernand Julien Dossa, Chang Ye, Jeff Braga, Dipam Chakraborty, Kinal Mehta, and João G.M. Araújo.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Evaluation Methods and Measures for Causal Learning Algorithms

IEEE Transactions on Artificial Intelligence

引用

IEEE Transactions on Artificial Intelligence 2022年第6期3卷 924-943页

作者： Cheng, Lu Guo, Ruocheng Moraffah, Raha Sheth, Paras Candan, K. Selcuk Liu, Huan Arizona State University School Of Computing Informatics And Decision Systems Engineering TempeAZ85287 United States City University Of Hong Kong School Of Data Science Hong Kong

The convenient access to copious multifaceted data has encouraged machine learning researchers to reconsider correlation-based learning and embrace the opportunity of causality-based learning, i.e., causal machine learning (causal learning). Recent years have, therefore, witnessed great effort in developing causal learning algorithms aiming to help artificial intelligence (AI) achieve human-level intelligence. Due to the lack of ground-truth data, one of the biggest challenges in current causal learning research is algorithm evaluations. This largely impedes the cross-pollination of AI and causal inference and hinders the two fields to benefit from the advances of the other. To bridge from conventional causal inference (i.e., based on statistical methods) to causal learning with Big Data (i.e., the intersection of causal inference and machine learning), in this survey, we review commonly used datasets, evaluation methods, and measures for causal learning using an evaluation pipeline similar to conventional machine learning. We focus on the two fundamental causal inference tasks and causality-aware machine learning tasks. Limitations of current evaluation procedures are also discussed. We, then, examine popular causal inference tools/packages and conclude with primary challenges and opportunities for benchmarking causal learning algorithms in the era of Big Data. The survey seeks to bring to the forefront the urgency of developing publicly available benchmarks and consensus-building standards for causal learning evaluation with observational data. In doing so, we hope to broaden the discussions and facilitate collaboration to advance the innovation and application of causal learning. © 2020 IEEE.

关键词： Maximum likelihood estimation

来源：评论

学校读者我要写书评

暂无评论

EMPATHIC: Emulating Human-Like Multimodal Personality Architecture Through Thoughtful Human-AI Conversation 14

EMPATHIC: Emulating Human-Like Multimodal Personality Archit...

引用

14th International Conference on Cloud computing, Data Science and engineering, Confluence 2024

作者： Varshini Devi, I. Oviya, I.R. Raja, Kalpana Amrita School of Computing Amrita Vishwa Vidyapeetham Department of Computer Science and Engineering Chennai India School of Medicine Yale University Section for Biomedical Informatics and Data Science New HavenCT United States

ISBN: (纸本)9798350344837

In the evolving landscape of human-computer interaction, this paper introduces an innovative framework poised to revolutionize chatbot systems. Our framework, meticulously designed for emotionally aware multimodal chatbots, seamlessly integrates real-time face and emotion recognition with natural language processing. Currently in the developmental phase, the framework exhibits a distinctive proficiency in executing a diverse array of human instructions. These instructions span tasks such as generating detailed captions, object counting, and responding to general queries. Leveraging parameter-efficient fine-tuning from OpenFlamingo, augmented by the inclusion of the Low-rank Adapter (LoRA), our framework strategically prioritizes versatility and operational efficiency. The core of our approach lies in the establishment of instruction templates that intricately merge both vision and language data. This strategic integration forms the backbone of our multi-modality instruction tuning methodology, enhancing the adaptability of the chatbot. Recognizing the pivotal role of high-quality training data, we emphasize its significance in shaping the dialogue performance of the chatbot. As the framework progresses through its developmental stages, this research lays a robust foundation for the creation of a versatile conversational agent. The envisioned applications of this agent range from enhancing customer service experiences to providing valuable support in mental health scenarios. © 2024 IEEE.

关键词： Emotion Recognition

来源：评论

学校读者我要写书评

暂无评论

A Lean-Congruence Format for EP-Bisimilarity 30

A Lean-Congruence Format for EP-Bisimilarity

引用

Combined 30th International Workshop on Expressiveness in Concurrency and 20th Workshop on Structural Operational Semantic, EXPRESS/SOS 2023

作者： van Glabbeek, Rob Höfner, Peter Wang, Weiyou School of Informatics University of Edinburgh United Kingdom School of Computer Science and Engineering University of New South Wales Sydney Australia School of Computing Australian National University Canberra Australia

Enabling preserving bisimilarity is a refinement of strong bisimilarity that preserves safety as well as liveness properties. To define it properly, labelled transition systems needed to be upgraded with a successor relation, capturing concurrency between transitions enabled in the same state. We enrich the well-known De Simone format to handle inductive definitions of this successor relation. We then establish that ep-bisimilarity is a congruence for the operators, as well as lean congruence for recursion, for all (enriched) De Simone languages. © R.J. van Glabbeek, P. Höfner & W. Wang.

关键词：

来源：评论

学校读者我要写书评

暂无评论

ChatGPT Rates Natural Language Explanation Quality Like Humans: But on Which Scales? 30

ChatGPT Rates Natural Language Explanation Quality Like Huma...

引用

Joint 30th International Conference on Computational Linguistics and 14th International Conference on Language Resources and Evaluation, LREC-COLING 2024

作者： Huang, Fan Kwak, Haewoon Park, Kunwoo An, Jisun Luddy School of Informatics Computing and Engineering Indiana University BloomingtonIN United States School of AI Convergence Soongsil University Seoul Korea Republic of Department of Intelligent Semiconductors Soongsil University Seoul Korea Republic of

ISBN: (纸本)9782493814104

As AI becomes more integral in our lives, the need for transparency and responsibility grows. While natural language explanations (NLEs) are vital for clarifying the reasoning behind AI decisions, evaluating them through human judgments is complex and resource-intensive due to subjectivity and the need for fine-grained ratings. This study explores the alignment between ChatGPT and human assessments across multiple scales (i.e., binary, ternary, and 7-Likert scale). We sample 300 data instances from three NLE datasets and collect 900 human annotations for both informativeness and clarity scores as the text quality measurement. We further conduct paired comparison experiments under different ranges of subjectivity scores, where the baseline comes from 8,346 human annotations. Our results show that ChatGPT aligns better with humans in more coarse-grained scales. Also, paired comparisons and dynamic prompting (i.e., providing semantically similar examples in the prompt) improve the alignment. This research advances our understanding of large language models' capabilities to assess the text explanation quality in different configurations for responsible AI development. © 2024 ELRA Language Resource Association: CC BY-NC 4.0.

关键词： Computational linguistics

来源：评论

学校读者我要写书评

暂无评论

Work-in-Progress: Security of Public-Key Schemes in the Quantum computing Era - A Literature Review

Work-in-Progress: Security of Public-Key Schemes in the Quan...

引用

2023 IEEE International Black Sea Conference on Communications and Networking, BlackSeaCom 2023

作者： Redkins, Boriss Kuzminykh, Ievgeniia Ghita, Bogdan King's College London Department of Informatics London United Kingdom School of Engineering Computing and Mathematics University of Plymouth Plymouth United Kingdom

ISBN: (纸本)9798350337822

Quantum computers, although still in the theoretical stage, conceptually have the power to break many current cryptosystems, including public-key schemes., for example, RSA (Rivest, Shamir., Adleman), elliptic-curve cryptography (ECC), and ElGamal. NIST specialists and national governments believe that existing cryptography algorithms will become vulnerable to brute force analysis by the end of the decade and agree that measures should be taken before then to ensure the secrecy of current and future algorithms. In this context, information security, including confidentiality and integrity, must be considered. This paper provides a systematic literature review on whether existing public-key schemes have the necessary complexity and solution space to resist a quantum computer. The discussion will focus on a quantum computer's negative 'real-world' impact and examine existing post-quantum algorithms, then provide analysis across two selected dimensions. Finally, we summarise the latest trends and gaps in post-quantum cryptography and provide the security analysis of NIST's last round finalists. © 2023 IEEE.

关键词： Public key cryptography

来源：评论

学校读者我要写书评

暂无评论

Bridging the Skills Gap: The Design Education Landscape with AI and Experiential Learning

引用

Design Management Journal 2024年第1期19卷 19-30页

作者： Pamela Napier Terri Wada Indiana University Herron School of Art and Design Indiana University Luddy School of Informatics Computing and Engineering Collabo XD

The rapid integration of artificial intelligence (AI) like ChatGPT into educational settings has irreversibly transformed teaching and learning paradigms. Recognizing that conventional instructional methods may fall short in equipping students with essential skills, this paper advocates for a shift towards experiential learning through service-learning (SL) and problem-based learning (PBL) in design education. These pedagogies, which promote intellectual growth and hands-on experience, are critical for students' career readiness and provide valuable real-world client engagement. Despite enthusiasm, incorporating SL and PBL faces significant hurdles, including a lack of educator training, resources, and structural support, coupled with the complexities of establishing partnerships, particularly affecting small businesses and disenfranchised entities. This research delves into the experiential learning landscape, examining successes and identifying skills gaps impacting educators and students. The findings, based on various design schools within a single university, assess challenges of integrating SL and PBL into design curricula and propose solutions to mitigate these barriers. Aiming to forge a stronger bridge between research, business, and academia, this paper underscores the necessity of preparing students for the 21st century's demands. Through this exploration, it contributes to the discourse on experiential learning within design education, enhancing the overall quality and relevance of student learning experiences.

关键词： Artificial Intelligence (AI) design education experiential learning problem-based learning service-learning

来源：评论

学校读者我要写书评

暂无评论

Neural Feature Predictor and Discriminative Residual Coding for Low-Bitrate Speech Coding 48

Neural Feature Predictor and Discriminative Residual Coding ...

引用

48th IEEE International Conference on Acoustics, Speech and Signal Processing, ICASSP 2023

作者： Yang, Haici Lim, Wootaek Kim, Minje Indiana University Luddy School of Informatics Computing and Engineering BloomingtonIN United States Electronics and Telecommunications Research Institute Daejeon34129 Korea Republic of

ISBN: (纸本)9781728163277

Low and ultra-low-bitrate neural speech codecs achieved unprecedented coding gain by generating speech signals from compact features. This paper introduces additional coding efficiency in speech coding by reducing the temporal redundancy existing in the frame-level feature sequence via a feature predictor. This predictor produces low-entropy residual representations, and we discriminatively code them based on their contribution to the signal reconstruction. Combining feature prediction and discriminative coding optimizes bitrate efficiency by assigning more bits to hard-to-predict events. We demonstrate the advantage of the proposed methods using the LPCNet as a neural vocoder, resulting in a scalable, lightweight, low-latency, and low-bitrate neural speech coding system. While our approach guarantees strict causality in the frame-level prediction, the subjective tests and feature space analysis show that our model achieves superior coding efficiency compared to the loosely-causal LPCNet and Lyra V2 in the very low bitrates. © 2023 IEEE.

关键词： Generative Model Low-bitrate Speech Codec LPCNet Predictive Coding

来源：评论

学校读者我要写书评

暂无评论

Paddy Plant Stress Identification Using Few-Shot Learning Framework 14

Paddy Plant Stress Identification Using Few-Shot Learning Fr...

引用

14th International Conference on Computer and Knowledge engineering, ICCKE 2024

作者： Moung, Ervin Gubin Sufian, Maisarah Mohd Farzamnia, Ali Naiidu, Pavindrah Naidu Narayanasamy Liaw, Valentino Angeline, Lorita Faculty of Computing and Informatics University Malaysia Sabah Sabah Kota Kinabalu Malaysia School of Computing and Engineering University of Huddersfield West YorkshireHD1 3DH United Kingdom Faculty of Engineering Universiti Malaysisa Sabah Sabah Kota Kinabalu Malaysia

ISBN: (纸本)9798331511272

The efficient identification of stress in paddy plant leaves is paramount for optimizing agricultural resources and ensuring robust crop yields in smart agriculture practices. This research explores the potential of few-shot learning (FSL) to address the inherent challenges posed by limited training data in stress identification. Three distinct FSL approaches - Siamese network, Matching network, and Model-Agnostic Meta Learning (MAML) - are evaluated for their accuracy in stress detection. The study introduces the importance of stress detection in smart agriculture and the challenges of limited training data. It then covers the methodology in five stages: dataset description, data pre-processing, FSL implementation, accuracy evaluation, and reporting. Among the FSL models, the Matching Network stands out with an impressive accuracy of 86% for 6-way 1-shot learning. This surpasses the performance of a Convolutional Neural Network (CNN) tested with a larger shot size (270-shots), which achieved an accuracy of 81%. These comparative results underscore the potential of FSL techniques in achieving precise stress identification, even when working with limited data. The objective of this research is to contribute valuable insights towards enhancing the efficiency of stress prediction in paddy leaves, thereby fostering healthier and more productive paddy crop production in smart agriculture. The findings presented here aim to inform the development of effective stress detection systems and advance the field of precision agriculture. ©2024 IEEE.

关键词： Zero-shot learning

来源：评论

学校读者我要写书评

暂无评论

Detecting cultural identity via robotic sensor data to understand differences during human-robot interaction

引用

Advanced Robotics 2023年第22期37卷 1446-1459页

作者： Lee, Jinjae Bennett, Casey C. Stanojevic, Cedomir Kim, Seongcheol Henkel, Zachary Baugus, Kenna Piatt, Jennifer A. Bethel, Cindy Sabanovic, Selma Department of Intelligence Computing Hanyang University Seoul Korea Republic of Department of Computing & Digital Media DePaul University ChicagoIL United States Department of Parks Recreation & Tourism Management Clemson University ClemsonSC United States Department of Computer Science and Engineering Mississippi State University Starkville StarkvilleMS United States School of Public Health Indiana University BloomingtonIN United States School of Informatics Computing and Engineering Indiana University BloomingtonIN United States

Socially-assistive robots (SARs) have significant potential to help manage chronic diseases (e.g. dementia, depression, diabetes) in spaces where people live, averse to clinic-based care. However, the challenge is designing SARs so that they perform appropriate interactions with people who have different characteristics, such as age, gender, and cultural identity. Those characteristics impact how human behaviors are performed as well as user expectations of robot responses. Although cross-cultural studies with robots have been conducted to understand differing population characteristics, they have mainly focused on statistical comparisons of groups. In this study, we utilize deep learning (DL) and machine learning (ML) models to evaluate whether cultural differences show up in robotic sensor data during human-robot interaction (HRI). To do so, a SAR was distributed to user's homes for three weeks in the US and Korea (25 participants), while collecting data on the human activity and the surrounding environment through on-board sensor devices. DL models based on that data were able to predict the user’s cultural identity with roughly 95% accuracy. Such findings have potential implications for the design and development of culturally-adaptive SARs to provide services across diverse cultural locales and multi-cultural environments where users’ cultural background cannot be assumed a priori. © 2023 Informa UK Limited, trading as Taylor & Francis Group and The Robotics Society of Japan.

关键词： Human robot interaction

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：