检索结果-内蒙古大学图书馆

26th European Conference on Artificial Intelligence, ECAI 2023

作者： Kong, Zhifeng Alfeld, Scott Computer Science and Engineering Uc San Diego La Jolla CA United States Computer Science Amherst College AmherstMA United States

ISBN: (纸本)9781643684369

Users have the right to have their data deleted by third-party learned systems, as codified by recent legislation such as the General Data Protection Regulation (GDPR) and the California Consumer Privacy Act (CCPA). Such data deletion can be achieved by full re-training, but this incurs a high computational cost for modern machine learning methods. To avoid this cost, many approximate deletion methods have been developed for supervised learning. Unsupervised learning, in contrast, remains largely an open problem when it comes to efficient approximate data deletion. In this paper, we introduce (1) an efficient method for approximate deletion in generative models, and (2) statistical tests for estimating whether training points have been deleted. We provide theoretical guarantees under various learner assumptions. We then empirically demonstrate our methods across a variety of generative methods. © 2023 The Authors.

关键词： Machine learning

来源：评论

学校读者我要写书评

暂无评论

FigurativeQA: A Test Benchmark for Figurativeness Comprehension for Question Answering 3

FigurativeQA: A Test Benchmark for Figurativeness Comprehens...

引用

3rd Workshop on Figurative Language Processing, FigLang 2022, as part of EMNLP 2022

作者： Rakshit, Geetanjali Flanigan, Jeffrey Computer Science and Engineering Department UC Santa Cruz United States

ISBN: (纸本)9781959429111

Figurative language is widespread in human language (Lakoff and Johnson, 2008), posing potential challenges in NLP applications. In this paper, we investigate the effect of figurative language on the task of question answering (QA). We construct FigurativeQA, a test set of 400 yes-no questions with figurative and non-figurative contexts, extracted from product reviews and restaurant reviews. We demonstrate that a state-of-the-art RoBERTa QA model has considerably lower performance in question answering when the contexts are figurative rather than literal, indicating a gap in current models. We propose a general method for improving the performance of QA models by converting the figurative contexts into non-figurative by prompting GPT-3, and demonstrate its effectiveness. Our results indicate a need for building QA models infused with figurative language understanding capabilities. © 2022 Association for Computational Linguistics.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Efficient Host Intrusion Detection using Hyperdimensional Computing

Efficient Host Intrusion Detection using Hyperdimensional Co...

引用

2024 IEEE International Conference on Big Data, BigData 2024

作者： Nam, Yujin King, Rachel Burke, Quinn Zhou, Minxuan McDaniel, Patrick Rosing, Tajana Uc San Diego Dept. of Computer Science and Engineering La Jolla United States University of Wisconsin-Madison Dept. of Computer Sciences Madison United States Illinois Institute of Technology Computer Science Department Chicago United States

ISBN: (纸本)9798350362480

Modern host-based intrusion detection systems (HIDS) rely on querying provenance graphs - graph representations of activity history on a system - to detect and respond to security threats present on a system. However, as the complexity and number of applications running on a system increase, the size of provenance graphs also increase, and thus the latency to query them. State-of-the-art designs deliver query latencies that are impractical for modern threat detection. In this paper, we introduce a hyper-dimensional computing (HDC) approach to querying provenance graphs for HIDS. By encoding provenance graphs and attack patterns/signatures into hyper-dimensional vectors, we can implement a query engine using simple vector operations. Our approach is hardware accelerator compatible, providing further speedups under resource-constrained environments. Our evaluation on a real-world dataset shows that our approach achieves > 90% detection accuracy and up to 4, 242× speedups over the state-of-the-art. This shows that HDC-based approaches can effectively deal with scaling issues in modern HIDS. © 2024 IEEE.

关键词： Network intrusion

来源：评论

学校读者我要写书评

暂无评论

Design and Development of a LMS with MVVM Architecture

Design and Development of a LMS with MVVM Architecture

引用

2023 International Conference on Sustainable Emerging Innovations in engineering and Technology, ICSEIET 2023

作者： Negi, Anurag Singh Vaidya, Himadri Chauhan, Rahul School of Computing Science & Engineering Galgotias University Greater Noida India School of Computer Science & Engineering Galgotias University Greater Noida India School of Computer Science Institute of Management Studies UC Campus Ghaziabad India

ISBN: (纸本)9798350329186

A learning management system promotes elearning and can also be used in open distance e-learning. In recent times, the number of mobile devices being used by peers has increased significantly, and an LMS app has thus gained increased importance in the field of E-learning and distance learning. Moodle is another application that is being used in universities, and in this paper, there is an app based on learning management as Moodle is more on the web side, and this app will be based on MVVM architecture, which is Model View. This paper will thus provide insights into the design and development of LMS app and provide an extensive survey of development process. © 2023 IEEE.

关键词： Architecture

来源：评论

学校读者我要写书评

暂无评论

Full-Distance Evasion of Pedestrian Detectors in the Physical World 38

Full-Distance Evasion of Pedestrian Detectors in the Physica...

引用

38th Conference on Neural Information Processing Systems, NeurIPS 2024

作者： Cheng, Zhi Hu, Zhanhao Liu, Yuqiu Li, Jianmin Su, Hang Hu, Xiaolin Department of Computer Science and Technology Tsinghua University Beijing China Department of Electrical Engineering and Computer Sciences UC Berkeley United States Department of Technology Beijing Forestry University Beijing China

Many studies have proposed attack methods to generate adversarial patterns for evading pedestrian detection, alarming the computer vision community about the need for more attention to the robustness of detectors. However, adversarial patterns optimized by these methods commonly have limited performance at medium to long distances in the physical world. To overcome this limitation, we identify two main challenges. First, in existing methods, there is commonly an appearance gap between simulated distant adversarial patterns and their physical world counterparts, leading to incorrect optimization. Second, there exists a conflict between adversarial losses at different distances, which causes difficulties in optimization. To overcome these challenges, we introduce a Full Distance Attack (FDA) method. Our physical world experiments demonstrate the effectiveness of our FDA patterns across various detection models like YOLOv5, Deformable-DETR, and Mask RCNN. Codes available at https://***/zhicheng2T0/***. © 2024 Neural information processing systems foundation. All rights reserved.

关键词：

来源：评论

学校读者我要写书评

暂无评论

The Future Is Rosie?: Disempowering Arguments about Automation and What to Do about It 25

The Future Is Rosie?: Disempowering Arguments about Automati...

引用

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

作者： Riek, Laurel D. Irani, Lilly Computer Science and Engineering UC San Diego San Diego CA United States Department of Communication UC San Diego San Diego CA United States

ISBN: (纸本)9798400713941

Many technologists who work in robotics and AI bristle at the idea that human worker displacement is problematic. Others wish to account for workers' needs, but face pervasive myths about the impacts of these technologies. This paper aims to clear the air by refuting five common arguments for automation: 1) the jobs being automated are undesirable, 2) labor shortages necessitate automation, 3) by "augmenting rather than automating"labor displacement will be prevented, 4) there will be new and better job creation, and 5) automation will give us all more leisure time. The advent of foundational models has led to an industrial gold rush, accelerating deployment without careful consideration of responsible and sustaiable design and deployment of these technologies. Despite technologists' best intentions, this path of pervasive automation we are on is not a good one, and we offer suggestions for how technologists, designers, and decision makers can push for worker-centered technological change moving forward. © 2025 Copyright held by the owner/author(s).

关键词： AI automation critical computing ethics future of work labor robotics social theory

来源：评论

学校读者我要写书评

暂无评论

ENTS: Flush-and-Fence-Free Failure Atomic Transactions 23

ENTS: Flush-and-Fence-Free Failure Atomic Transactions

引用

Proceedings of the International Symposium on Memory Systems

作者： Yun Joon Soh Steven Swanson Jishen Zhao Computer Science & Engineering UC San Diego USA

ISBN: (纸本)9798400716447

Persistent memory (PMEM) offers applications with DRAM-like performance and non-volatility. Yet programmers must ensure crash consistency — guaranteeing data consistency upon a sudden power failure — when implementing PMEM programs by carefully placing flush and fence instructions. That often introduces performance overhead and increases the risk of bugs. Previous approaches to reducing flush and fence instructions either require extensive hardware modifications or are only partially successful. We propose Extricated Non-Temporal Store (ENTS), a programming library that removes explicit flush and fence instructions and provides an easy-to-use and low-overhead interface for failure atomic transactional PMEM programs without hardware modification. ENTS achieves crash consistency with a novel technique, Persist-On-Write (POW), which issues a fence-free non-temporal store (cache bypassing store) on each volatile data object upon modification. Instead of an explicit fence instruction, ENTS leverages fence-like instructions (e.g., lock or xchg under x86_64), which are already prevalent in the failure atomic transactional programs’ concurrency control mechanisms. We evaluate ENTS on seven workloads: four data structures (B+Tree, RBTree, Hashmap, Skiplist), two transaction benchmarks (TPC-C, TATP), and a write-optimized hashtable (Level-Hash). Programs gain 1.8 × and 2.1 × higher throughput than Clobber-NVM (a compiler-directed crash-consistency tool) and PMDK (an industry-standard library), respectively.

关键词： Crash Consistency Failure Atomic Transaction Persistent Memory

来源：评论

学校读者我要写书评

暂无评论

Private and Efficient Learning With Hyperdimensional Computing

IEEE Transactions on Circuits and Systems for Artificial Int...

引用

IEEE Transactions on Circuits and Systems for Artificial Intelligence 2024年第2期1卷 204-219页

作者： Behnam Khaleghi Xiaofan Yu Jaeyoung Kang Xuan Wang Tajana Rosing Department of Computer Science and Engineering UC San Diego La Jolla CA USA

Machine learning algorithms, especially deep neural networks (DNNs), are computationally expensive and vulnerable to privacy breaches. Hyperdimensional Computing (HDC) is a brain-inspired machine learning paradigm which offers a potential alternative for efficient and private learning. HDC encodes dense, sensory inputs into sparse, high-dimensional vectors of 10,000 bits or more, and learns using well-defined and lightweight operations on these vectors. However, naïve HDC is vulnerable to privacy attacks as the encoding process can be reversely determined. In this paper, we systematically study the fundamental privacy challenges of HDC due to reversibility and propose an end-to-end private training and inference framework for HDC with efficient hardware implementation. We first show that HDC is not naturally private as the original data and encoding parameters can be reconstructed from the encoded vector, regardless of the encoding methods applied. We term this attack as the feature extraction attack. We then propose a locally sparse encoding method to obfuscate the encoded information for protecting inference privacy, leading to computational and communication savings. We further propose Privé-HDnn which uses a hybrid architecture of Convolutional Neural Network (CNN) as feature extractors and private-preserved HDC classifier for complex tasks such as image classification. Thanks to the noise tolerance of HDC, Privé-HDnn enables private single- and multi-pass training of HDC by differential privacy-guaranteed noise injection. Results show that our new encoding method defends against decoding attacks with an RMSE deviation of 0.42, while also speeding up encoding by 2.6–5.3 $\times$ (19–79 $\times$ ) compared to the FPGA (CPU) baseline. For privacy-preserved training, Privé-HDnn achieves comparable or up to 5.8% better accuracy than DNNs with up to 1231 $\times$ faster training time on GPU as compared to the state-of-the-art NNs.

关键词： Encoding Vectors Training Privacy Feature extraction Noise Iterative methods Field programmable gate arrays Hardware Root mean square

来源：评论

学校读者我要写书评

暂无评论

Forecasting Market Price of Stock Using Artificial Neural Network

Forecasting Market Price of Stock Using Artificial Neural Ne...

引用

2023 International Conference on Sustainable Emerging Innovations in engineering and Technology, ICSEIET 2023

作者： Srivastava, Yuvraj Chauhan, Alok Singh Vaishnav, Rashmi Bharadwaj, Kapil School of Computing Science & Engineering Galgotias University Greater Noida India School of Computer Science Institute of Management Studies UC Campus Ghaziabad India

ISBN: (纸本)9798350329186

Stocks represent the ownership stake in a company and reflect its financial performance and contributions to the economy. Company stocks are divided into shares. Analysing stock market decisions is a complex process involving factors such as price trends, market demand, company stability, rumours, brand reputation, and venture capital. It is crucial to have access to vital information that allows investors to understand and assess the associated risks. While various methods like technical analysis, time series analysis, and statistical analysis attempt to predict stock prices, none of these methods are consistently reliable. Therefore, artificial neural networks, a branch of Artificial Intelligence, offer a promising approach to identify both known and hidden patterns in data. The process involves two phases: training and prediction. The Backpropagation algorithm is utilized for the training session, while a Multilayer Feedforward Network serves as the model for price prediction. This forecasting is based on multiple parameters derived from company data. © 2023 IEEE.

关键词： Neural networks

来源：评论

学校读者我要写书评

暂无评论

CoLLAP: Contrastive Long-form Language-Audio Pretraining with Musical Temporal Structure Augmentation

CoLLAP: Contrastive Long-form Language-Audio Pretraining wit...

引用

International Conference on Acoustics, Speech, and Signal Processing (ICASSP)

作者： Junda Wu Warren Li Zachary Novack Amit Namburi Carol Chen Julian McAuley Computer Science and Engineering UC San Diego La Jolla USA Computer Science Department UC Los Angeles Los Angeles USA

ISBN: (数字)9798350368741

ISBN: (纸本)9798350368758

Modeling temporal characteristics plays a significant role in the representation learning of audio waveform. We propose Contrastive Long-form Language-Audio Pretraining (CoLLAP) to significantly extend the perception window for both the input audio (up to 5 minutes) and the language descriptions (exceeding 250 words), while enabling contrastive learning across modalities and temporal dynamics. Leveraging recent Music-LLMs to generate long-form music captions for full-length songs, augmented with musical temporal structures, we collect 51.3K audio-text pairs derived from the large-scale AudioSet training dataset, where the average audio length reaches 288 seconds. We propose a novel contrastive learning architecture that fuses language representations with structured audio representations by segmenting each song into clips and extracting their embeddings. With an attention mechanism, we capture multimodal temporal correlations, allowing the model to automatically weigh and enhance the final fusion score for improved contrastive alignment. Finally, we develop two variants of the CoLLAP model with different types of backbone language models. Through comprehensive experiments on multiple long-form music-text retrieval datasets, we demonstrate consistent performance improvement in retrieval accuracy compared with baselines. We also show the pretrained CoLLAP models can be transferred to various music information retrieval tasks, with heterogeneous long-form multimodal contexts.

关键词： Representation learning Training Attention mechanisms Correlation Fuses Contrastive learning Signal processing Speech processing Music information retrieval Context modeling

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：