检索结果-内蒙古大学图书馆

arXiv 2024年

作者： Pons, Joan Carles López, Pau Vives Murakami, Yukihiro van Iersel, Leo Department of Mathematics and Computer Science Universitat de les Illes Balears Spain Delft Institute of Applied Mathematics Delft University of Technology Netherlands

Reticulate evolution can be modelled using phylogenetic networks. Tree-based networks, which are one of the more general classes of phylogenetic networks, have recently gained eminence for its ability to represent evolutionary histories with an underlying tree structure. To better understand tree-based networks, numerous characterizations have been proposed, based on tree embeddings, matchings, and arc partitions. Here, we build a bridge between two arc partition characterizations, namely maximal fence decompositions and cherry covers. Results on cherry covers have been found for general phylogenetic networks. We first show that the number of cherry covers is the same as the number of support trees (underlying tree structure of tree-based networks) for a given semi-binary network. Maximal fence decompositions have only been defined thus far for binary networks (constraints on vertex degrees). We remedy this by generalizing fence decompositions to non-binary networks, and using this, we characterize semi-binary tree-based networks in terms of forbidden structures. Furthermore, we give an explicit enumeration of cherry covers of semi-binary networks, by studying its fence decomposition. Finally, we prove that it is possible to characterize semi-binary tree-child networks, a subclass of tree-based networks, in terms of the number of their cherry covers. Copyright © 2024, The Authors. All rights reserved.

关键词： Fences

来源：评论

学校读者我要写书评

暂无评论

Aligning to What? Limits to RLHF Based Alignment

arXiv

引用

arXiv 2025年

作者： Barnhart, Logan Bafghi, Reza Akbarian Becker, Stephen Raissi, Maziar Department of Applied Mathematics University of Colorado Boulder United States Department of Computer Science University of Colorado Boulder United States Department of Mathematics University of California Riverside United States

Reinforcement Learning from Human Feedback (RLHF) is increasingly used to align large language models (LLMs) with human preferences. However, the effectiveness of RLHF in addressing underlying biases remains unclear. This study investigates the relationship between RLHF and both covert and overt biases in LLMs, particularly focusing on biases against African Americans. We applied various RLHF techniques (DPO, ORPO, and RLOO) to Llama 3 8B and evaluated the covert and overt biases of the resulting models using matched-guise probing and explicit bias testing. We performed additional tests with DPO on different base models and datasets;among several implications, we found that SFT before RLHF calcifies model biases. Additionally, we extend the tools for measuring biases to multi-modal models. Through our experiments we collect evidence that indicates that current alignment techniques are inadequate for nebulous tasks such as mitigating covert biases, highlighting the need for capable datasets, data curating techniques, or alignment tools.1 © 2025, CC BY.

关键词： Reinforcement learning

来源：评论

学校读者我要写书评

暂无评论

Local Sequential MCMC for Data Assimilation with Applications in Geoscience

arXiv

引用

arXiv 2024年

作者： Ruzayqat, Hamza Knio, Omar Applied Mathematics and Computational Science Program Computer Electrical and Mathematical Sciences and Engineering Division King Abdullah University of Science and Technology Thuwal23955-6900 Saudi Arabia

This paper presents a new data assimilation (DA) scheme based on a sequential Markov Chain Monte Carlo (SMCMC) DA technique [36] which is provably convergent and has been recently used for filtering, particularly for high-dimensional non-linear, and potentially, non-Gaussian state-space models. Unlike particle filters, which can be considered exact methods and can be used for filtering non-linear, non-Gaussian models, SMCMC does not assign weights to the samples/particles, and therefore, the method does not suffer from the issue of weight-degeneracy when a relatively small number of samples is used. We design a localization approach within the SMCMC framework that focuses on regions where observations are located and restricts the transition densities included in the filtering distribution of the state to these regions. This results in immensely reducing the effective degrees of freedom and thus improving the efficiency. We test the new technique on high-dimensional (d ∼ 104 − 105) linear Gaussian model and non-linear shallow water models with Gaussian noise with real and synthetic observations. For two of the numerical examples, the observations mimic the data generated by the Surface Water and Ocean Topography (SWOT) mission led by NASA, which is a swath of ocean height observations that changes location at every assimilation time step. We also use a set of ocean drifters' real observations in which the drifters are moving according the ocean kinematics and assumed to have uncertain locations at the time of assimilation. We show that when higher accuracy is required, the proposed algorithm is superior in terms of efficiency and accuracy over competing ensemble methods and the original SMCMC *** Codes 62M20, 60G35, 60J20, 94A12, 93E11, 65C40 © 2024, CC BY.

关键词： Markov chains

来源：评论

学校读者我要写书评

暂无评论

Securing IoT Firmware Dispatch Systems with Blockchain

Securing IoT Firmware Dispatch Systems with Blockchain

引用

IEEE International Conference on Blockchain (Blockchain)

作者： Vince Biró Wei-Yang Chiu Weizhi Meng Department of Applied Mathematics and Computer Science SPTAGE Lab Technical University of Denmark Denmark

Internet-of-Things (IoT) devices have become a prevalent part of everyday life, making the need for IoT firmware security greater than ever. Firmware dispatch systems play a crucial role in maintaining the secure operations of IoT devices, where a compromise on the integrity of dispatches may simply ruin an IoT device. Hence it is essential to have robust solutions to safeguard such systems. With the advent of blockchain and Web3 technologies, there are new options for tackling security-related challenges. Blockchain provides a tamper-proof, secure, and fully distributed protocol, making it an excellent choice for a firmware dispatch system. In this work, we explore the existing blockchain-based solutions for distributing IoT firmware updates, and then propose a secure firmware dispatch system for IoT devices using Ethereum. Our system employs ***, with a hosted implementation of the InterPlanetary File System (IPFS) and a peer-to-peer hypermedia protocol. In the evaluation, we demonstrate that our designed architecture is fast, reliable, and scalable by offering a firmware integrity validation. It can benefit a wide range of manufacturers who intend to increase the security of their IoT devices.

关键词：

来源：评论

学校读者我要写书评

暂无评论

Low Acceptance Agreement Tests via Bounded-Degree Symplectic HDXs

Low Acceptance Agreement Tests via Bounded-Degree Symplectic...

引用

Annual IEEE Symposium on Foundations of computer science

作者： Yotam Dikstein Irit Dinur Alexander Lubotzky Computer Science and Discrete Mathematics Institute for Advanced Study Princeton New Jersey USA Dept. of Applied Math and Computer Science The Weizmann Institute of Science Rehovot Israel Department of Mathematics The Weizmann Institute of Science Rehovot Israel

ISBN: (数字)9798331516741

ISBN: (纸本)9798331516758

We solve the derandomized direct product testing question in the low acceptance regime, by constructing new high dimensional expanders that have no small connected covers. We show that our complexes have swap cocycle expansion, which allows us to deduce the agreement theorem by relying on previous work. Derandomized direct product testing, also known as agreement testing, is the following problem. Let $X$ be a family of k-element subsets of $[N]$ and let $\{f_{s}:s\rightarrow\Sigma\vert s\in X\}$ be an ensemble of local functions, each defined over a subset $s\subset\lceil N$ . Suppose that we run the following so-called agreement test: choose a random pair of sets $s_{1}, s_{2}\in X$ that intersect on $\sqrt{k}$ elements, and accept if $f_{s_{1}}, f_{s_{2}}$ agree on the elements in $s_{1}\cap s_{2}$ . We denote the success probability of this test by Agree $\{f_{s}\})$ Given that Agree $(\{f_{s}\})=\varepsilon > 0$ is there a global function $G:[N]\rightarrow\Sigma$ such that $f_{s}=G\vert _{s}$ for a non-negligible fraction of $s\in X\ ?$ We construct a family $X$ of k-subsets of $[N]$ such that $\vert X\vert =O(N)$ , and such that it satisfies the low acceptance agreement theorem. Namely, $\text{Agree}\left(\left\{f_s\right\}\right)>\varepsilon \Longrightarrow \exists G:[N] \rightarrow \Sigma, \quad \underset{s}{\mathbb{P}}\left[\left.f_s \stackrel{0.99}{\approx} G\right\vert_s\right] \geqslant \text{poly}(\varepsilon)$ . A key idea is to replace the well-studied LSV complexes by symplectic high dimensional expanders (HDXs). The family $X$ is just the k-faces of the new symplectic HDXs. The latter serve our needs better since their fundamental group satisfies the congruence subgroup property, which implies that they lack small covers. We also give a polynomial-time algorithm to construct this family of sym-plectic HDXs.

关键词： computer science Testing

来源：评论

学校读者我要写书评

暂无评论

An Adaptive Streaming Feature Selection Technique for Classifying Non-Stationary Data Streams 7

An Adaptive Streaming Feature Selection Technique for Classi...

引用

7th International Symposium on Multidisciplinary Studies and Innovative Technologies, ISMSIT 2023

作者： Arya, Monika Dewangan, Bhupesh Kumar Choudhury, Tanupriya Kotecha, Ketan Dewangan, Sanjana Gaidhane, Akhilesh Bhilai Institute of Technology Department of Computer Science and Engineering Durg India Op Jindal University Department of Computer Science and Engineering Raigarh496109 India Pune411045 India Symbiosis Centre for Applied Artificial Intelligence Pune411045 India School of Science Op Jindal University Raigarh India Sages M.G. Ward Department of Mathematics Kondagaon India

ISBN: (纸本)9798350342154

In today's digital era, various real-world applications generate data in streams, and these data streams are of two types stationary data streams, which are static, and non-stationary data streams that are dynamic in nature. Non-stationary data streams are pro-to-concept drifts. To overcome this issue, an adaptable streaming feature selection technique for non-stationary data streams is proposed in this paper. The selection of relevant feature space from non-stationary data streams is crucial for enabling intelligence through various data mining tasks, specifically using data classification. Feature selection improves the overall performance of the classifier while increasing the computational overhead. The proposed approach aims to enable streaming feature selection whenever the classification accuracy is underperforming to enhance the classification accuracy, while the model remains stable if there is no change in the concepts. The proposed approach is adaptive and ameliorates the classifier performance by reducing the error rate during changes in the concept with time. This adaptability to concept drift improves the accuracy by approximately 16% and reduces the computation time by 50%. The proposed model avoids overfitting of the model. © 2023 IEEE.

关键词： Feature Selection

来源：评论

学校读者我要写书评

暂无评论

Protecting User Privacy in Online Settings via Supervised Learning

arXiv

引用

arXiv 2023年

作者： Rusescu, Alexandru Lampe, Brooke Meng, Weizhi SPTAGE Lab Department of Applied Mathematics and Computer Science Technical University of Denmark Denmark

Companies that have an online presence—in particular, companies that are exclusively digital—often subscribe to this business model: collect data from the user base, then expose the data to advertisement agencies in order to turn a profit. Such companies routinely market a service as "free," while obfuscating the fact that they tend to "charge" users in the currency of personal information rather than money. However, online companies also gather user data for more principled purposes, such as improving the user experience and aggregating statistics. The problem is the sale of user data to third parties. In this work, we design an intelligent approach to online privacy protection that leverages supervised learning. By detecting and blocking data collection that might infringe on a user’s privacy, we can restore a degree of digital privacy to the user. In our evaluation, we collect a dataset of network requests and measure the performance of several classifiers that adhere to the supervised learning paradigm. The results of our evaluation demonstrate the feasibility and potential of our approach. © 2023, CC BY-NC-ND.

关键词： Decision trees

来源：评论

学校读者我要写书评

暂无评论

Exact Output Synchronization of Discrete-Time Multi-Agent Systems in the Presence of Disturbances and Measurement Noise with Known Frequencies via Scale-Free Protocol Design 42

Exact Output Synchronization of Discrete-Time Multi-Agent Sy...

引用

42nd Chinese Control Conference, CCC 2023

作者： Liu, Zhenwei Zhang, Meirong Saberi, Ali Stoorvogel, Anton A. College of Information Science and Engineering Northeastern University Shenyang110819 China School of Engineering and Applied Science Gonzaga University SpokaneWA United States School of Electrical Engineering and Computer Science Washington State University PullmanWA99164 United States Mathematics and Computer Science Department of Electrical Engineering University of Twente Enschede Netherlands

ISBN: (纸本)9789887581543

This paper aims to achieve scalable exact output synchronization for discrete-time homogeneous multi-agent systems in the presence of disturbances and measurement noise with known frequencies. The key contribution of this paper is on the distributed protocol that only uses the information of agent models, rather than the communication network information and the agent number, so as to achieve the scalable exact synchronization under disturbances and measurement noise. The validity of the protocol is verified by numerical simulations with arbitrarily chosen number of agents. © 2023 Technical Committee on Control Theory, Chinese Association of Automation.

关键词： Multi agent systems

来源：评论

学校读者我要写书评

暂无评论

Towards A Scalable and Privacy-Preserving Blockchain-based European Parking System

Towards A Scalable and Privacy-Preserving Blockchain-based E...

引用

International Conference on Parallel and Distributed Systems (ICPADS)

作者： Jonathan Kvist Brittain Wei-Yang Chiu Weizhi Meng Department of Applied Mathematics and Computer Science SPTAGE Lab Technical University of Denmark Denmark

The importance of efficient and accessible parking systems has been growing over the past decades. Steady growth in urban population and car ownership has increased problems with traffic congestion and air pollution caused by inadequate parking systems. Blockchain-based parking systems have been proposed to increase system availability and resilience and improve trust among participants. However, these systems are not transferable to the parking systems of European cities such as Copenhagen, as they are based on assumptions about the parking infrastructure, which do not hold, and are inherently incompatible with regional privacy protection regulations such as the GDPR. Furthermore, many blockchain solutions suffer from scalability issues, severely limiting their efficiency. In this work, we develop a blockchainbased parking system in Denmark, aiming to make up the gap in the existing research by directly considering GDPR compliance. Our work focuses on the municipal parking system for on-street parking in Copenhagen (Denmark), where Hyperledger Fabric is used to maintain a trusted distributed ledger for parking data shared by the network. Personal data is protected through offchain storage while maintaining on-chain verifiability. The proposed system is implemented as a proof-of-concept application, which can deliver sufficient throughput to support the needs of municipal parking.

关键词： Distributed ledger System performance Urban areas Europe Throughput Regulation Fabrics

来源：评论

学校读者我要写书评

暂无评论

BlockPAT: A Blockchain-Enabled Second-Hand Physical Asset Tokenization Management System

BlockPAT: A Blockchain-Enabled Second-Hand Physical Asset To...

引用

International Conference on Distributed Computing Systems

作者： Jiaxuan Wu Wei-Yang Chiu Weizhi Meng Brooke Lampe Department of Applied Mathematics and Computer Science SPTAGE Lab Technical University of Denmark Denmark

In this work, we develop BlockPAT, a blockchain-enabled management system for the tokenization of second-hand physical assets, e.g., laptops. With this system, the information gap between buyers and sellers in the second-hand market will be eliminated, and with the help of a price oracle, the liquidity of the second-hand market can be greatly improved. Furthermore, our system is built upon the latest ZK-rollups solution; thus, the overall transaction cost and time delay will be limited to an affordable value.

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：