帮我润色论文。In this paper, we propose a balancing training method to address problems in imbalanced data learning. To this end, we derive a new loss used in the balancing training phase that alleviates the influence of samples that cause an overfitted decision boundary. The proposed loss efficiently improves the performance of any type of imbalance learning methods. In experiments on multiple benchmark data sets, we demonstrate the validity of our method and reveal that the proposed loss outperforms the state-of-the-art cost-sensitive loss methods. Furthermore, since our loss is not restricted to a specific task, model, or training method, it can be easily used in combination with other recent resampling, meta-learning, and cost-sensitive learning methods for class-imbalance problems. Our code is made available at https://github.com/pseulki/IB-Loss.

时间: 2023-02-24 13:02:14 浏览: 323
PDF

Development of Fatigue Driving Detection Method Based on Fuzzy Control Logic

在本文中,我们提出了一种平衡训练方法,以解决不平衡数据学习中的问题。为此,我们在平衡训练阶段推导出一种新的损失,以减少导致过度拟合决策边界的样本的影响。所提出的损失有效提高了任何类型的不平衡学习方法的性能。在多个基准数据集的实验中,我们证明了我们的方法的有效性,并发现所提出的损失优于最先进的成本敏感损失方法。此外,由于我们的损失不受特定任务、模型或训练方法的限制,它可以轻松与其他最近的重采样、元学习和成本敏感学习方法结合使用,以解决类不平衡问题。我们的代码可在 https://github.com/pseulki/IB-Loss 获得。
阅读全文

相关推荐

The human visual cortex is biased towards shape components while CNNs produce texture biased features. This fact may explain why the performance of CNN significantly degrades with low-labeled input data scenarios. In this paper, we propose a frequency re-calibration U-Net (FRCU-Net) for medical image segmentation. Representing an object in terms of frequency may reduce the effect of texture bias, resulting in better generalization for a low data regime. To do so, we apply the Laplacian pyramid in the bottleneck layer of the U-shaped structure. The Laplacian pyramid represents the object proposal in different frequency domains, where the high frequencies are responsible for the texture information and lower frequencies might be related to the shape. Adaptively re-calibrating these frequency representations can produce a more discriminative representation for describing the object of interest. To this end, we first propose to use a channel-wise attention mechanism to capture the relationship between the channels of a set of feature maps in one layer of the frequency pyramid. Second, the extracted features of each level of the pyramid are then combined through a non-linear function based on their impact on the final segmentation output. The proposed FRCU-Net is evaluated on five datasets ISIC 2017, ISIC 2018, the PH2, lung segmentation, and SegPC 2021 challenge datasets and compared to existing alternatives, achieving state-of-the-art results.请详细介绍这段话中的技术点和实现方式

Algorithm 1: The online LyDROO algorithm for solving (P1). input : Parameters V , {γi, ci}Ni=1, K, training interval δT , Mt update interval δM ; output: Control actions 􏰕xt,yt􏰖Kt=1; 1 Initialize the DNN with random parameters θ1 and empty replay memory, M1 ← 2N; 2 Empty initial data queue Qi(1) = 0 and energy queue Yi(1) = 0, for i = 1,··· ,N; 3 fort=1,2,...,Kdo 4 Observe the input ξt = 􏰕ht, Qi(t), Yi(t)􏰖Ni=1 and update Mt using (8) if mod (t, δM ) = 0; 5 Generate a relaxed offloading action xˆt = Πθt 􏰅ξt􏰆 with the DNN; 6 Quantize xˆt into Mt binary actions 􏰕xti|i = 1, · · · , Mt􏰖 using the NOP method; 7 Compute G􏰅xti,ξt􏰆 by optimizing resource allocation yit in (P2) for each xti; 8 Select the best solution xt = arg max G 􏰅xti , ξt 􏰆 and execute the joint action 􏰅xt , yt 􏰆; { x ti } 9 Update the replay memory by adding (ξt,xt); 10 if mod (t, δT ) = 0 then 11 Uniformly sample a batch of data set {(ξτ , xτ ) | τ ∈ St } from the memory; 12 Train the DNN with {(ξτ , xτ ) | τ ∈ St} and update θt using the Adam algorithm; 13 end 14 t ← t + 1; 15 Update {Qi(t),Yi(t)}N based on 􏰅xt−1,yt−1􏰆 and data arrival observation 􏰙At−1􏰚N using (5) and (7). i=1 i i=1 16 end With the above actor-critic-update loop, the DNN consistently learns from the best and most recent state-action pairs, leading to a better policy πθt that gradually approximates the optimal mapping to solve (P3). We summarize the pseudo-code of LyDROO in Algorithm 1, where the major computational complexity is in line 7 that computes G􏰅xti,ξt􏰆 by solving the optimal resource allocation problems. This in fact indicates that the proposed LyDROO algorithm can be extended to solve (P1) when considering a general non-decreasing concave utility U (rit) in the objective, because the per-frame resource allocation problem to compute G􏰅xti,ξt􏰆 is a convex problem that can be efficiently solved, where the detailed analysis is omitted. In the next subsection, we propose a low-complexity algorithm to obtain G 􏰅xti, ξt􏰆. B. Low-complexity Algorithm for Optimal Resource Allocation Given the value of xt in (P2), we denote the index set of users with xti = 1 as Mt1, and the complementary user set as Mt0. For simplicity of exposition, we drop the superscript t and express the optimal resource allocation problem that computes G 􏰅xt, ξt􏰆 as following (P4) : maximize 􏰀j∈M0 􏰕ajfj/φ − Yj(t)κfj3􏰖 + 􏰀i∈M1 {airi,O − Yi(t)ei,O} (28a) τ,f,eO,rO 17 ,建立了什么模型

Another example is the SRIOV_NET_VF resource class, which is provided by SRIOV-enabled network interface cards. In the case of multiple SRIOV-enabled NICs on a compute host, different qualitative traits may be tagged to each NIC. For example, the NIC called enp2s0 might have a trait “CUSTOM_PHYSNET_PUBLIC” indicating that the NIC is attached to a physical network called “public”. The NIC enp2s1 might have a trait “CUSTOM_PHYSNET_INTRANET” that indicates the NIC is attached to the physical network called “Intranet”. We need a way of representing that these NICs each provide SRIOV_NET_VF resources but those virtual functions are associated with different physical networks. In the resource providers data modeling, the entity which is associated with qualitative traits is the resource provider object. Therefore, we require a way of representing that the SRIOV-enabled NICs are themselves resource providers with inventories of SRIOV_NET_VF resources. Those resource providers are contained on a compute host which is a resource provider that has inventory records for other types of resources such as VCPU, MEMORY_MB or DISK_GB. This spec proposes that nested resource providers be created to allow for distinguishing details of complex components of some resource providers. During review the question came up about “rolling up” amounts of these nested providers to the root level. Imagine this scenario: I have a NIC with two PFs, each of which has only 1 VF available, and I get a request for 2 VFs without any traits to distinguish them. Since there is no single resource provider that can satisfy this request, it will not select this root provider, even though the root provider “owns” 2 VFs. This spec does not propose any sort of “rolling up” of inventory, but this may be something to consider in the future. If it is an idea that has support, another BP/spec can be created then to add this behavior.

Recall that to solve (P2) in the tth time frame, we observe ξt 􏰗 {hti, Qi(t), Yi(t)}Ni=1, consisting of the channel gains {hti}Ni=1 and the system queue states {Qi(t),Yi(t)}Ni=1, and accordingly decide the control action {xt, yt}, including the binary offloading decision xt and the continuous resource allocation yt 􏰗 􏰄τit, fit, eti,O, rit,O􏰅Ni=1. A close observation shows that although (P2) is a non-convex optimization problem, the resource allocation problem to optimize yt is in fact an “easy” convex problem if xt is fixed. In Section IV.B, we will propose a customized algorithm to efficiently obtain the optimal yt given xt in (P2). Here, we denote G􏰀xt,ξt􏰁 as the optimal value of (P2) by optimizing yt given the offloading decision xt and parameter ξt. Therefore, solving (P2) is equivalent to finding the optimal offloading decision (xt)∗, where (P3) : 􏰀xt􏰁∗ = arg maximize G 􏰀xt, ξt􏰁 . (20) xt ∈{0,1}N In general, obtaining (xt)∗ requires enumerating 2N offloading decisions, which leads to significantly high computational complexity even when N is moderate (e.g., N = 10). Other search based methods, such as branch-and-bound and block coordinate descent [29], are also time-consuming when N is large. In practice, neither method is applicable to online decision- making under fast-varying channel condition. Leveraging the DRL technique, we propose a LyDROO algorithm to construct a policy π that maps from the input ξt to the optimal action (xt)∗, i.e., π : ξt 􏰕→ (xt)∗, with very low complexity, e.g., tens of milliseconds computation time (i.e., the time duration from observing ξt to producing a control action {xt, yt}) when N = 10.,为什么要使用深度强化学习

最新推荐

recommend-type

SCI论文写作框架及英文常用句型模板.pdf

- 描述方法时,可以采用:"We propose a new … model and … algorithm that enables us to …" 或 "In this paper, we describe improved … models for …"。 - 结果部分,例如:"This paper reports results ...
recommend-type

dnSpy-net-win32-222.zip

dnSpy-net-win32-222.zip
recommend-type

和美乡村城乡融合发展数字化解决方案.docx

和美乡村城乡融合发展数字化解决方案.docx
recommend-type

如何看待“适度宽松”的货币政策.pdf

如何看待“适度宽松”的货币政策.pdf
recommend-type

GitHub图片浏览插件:直观展示代码中的图像

资源摘要信息: "ImagesOnGitHub-crx插件" 知识点概述: 1. 插件功能与用途 2. 插件使用环境与限制 3. 插件的工作原理 4. 插件的用户交互设计 5. 插件的图标和版权问题 6. 插件的兼容性 1. 插件功能与用途 插件"ImagesOnGitHub-crx"设计用于增强GitHub这一开源代码托管平台的用户体验。在GitHub上,用户可以浏览众多的代码仓库和项目,但GitHub默认情况下在浏览代码仓库时,并不直接显示图像文件内容,而是提供一个“查看原始文件”的链接。这使得用户体验受到一定限制,特别是对于那些希望直接在网页上预览图像的用户来说不够方便。该插件正是为了解决这一问题,允许用户在浏览GitHub上的图像文件时,无需点击链接即可直接在当前页面查看图像,从而提供更为流畅和直观的浏览体验。 2. 插件使用环境与限制 该插件是专为使用GitHub的用户提供便利的。它能够在GitHub的代码仓库页面上发挥作用,当用户访问的是图像文件页面时。值得注意的是,该插件目前只支持".png"格式的图像文件,对于其他格式如.jpg、.gif等并不支持。用户在使用前需了解这一限制,以免在期望查看其他格式文件时遇到不便。 3. 插件的工作原理 "ImagesOnGitHub-crx"插件的工作原理主要依赖于浏览器的扩展机制。插件安装后,会监控用户在GitHub上的操作。当用户访问到图像文件对应的页面时,插件会通过JavaScript检测页面中的图像文件类型,并判断是否为支持的.png格式。如果是,它会在浏览器地址栏的图标位置上显示一个小octocat图标,用户点击这个图标即可触发插件功能,直接在当前页面上查看到图像。这一功能的实现,使得用户无需离开当前页面即可预览图像内容。 4. 插件的用户交互设计 插件的用户交互设计体现了用户体验的重要性。插件通过在地址栏中增加一个小octocat图标来提示用户当前页面有图像文件可用,这是一种直观的视觉提示。用户通过简单的点击操作即可触发查看图像的功能,流程简单直观,减少了用户的学习成本和操作步骤。 5. 插件的图标和版权问题 由于插件设计者在制作图标方面经验不足,因此暂时借用了GitHub的标志作为插件图标。插件的作者明确表示,如果存在任何错误或版权问题,将会进行更改。这体现了开发者对知识产权尊重的态度,同时也提醒了其他开发者在使用或设计相关图标时应当考虑到版权法律的约束,避免侵犯他人的知识产权。 6. 插件的兼容性 插件的兼容性是评估其可用性的重要标准之一。由于插件是为Chrome浏览器的用户所设计,因此它使用了Chrome扩展程序的标准格式,即.crx文件。用户需要通过浏览器的扩展程序管理界面进行安装。尽管目前插件仅支持.png图像格式,但对于希望在GitHub上浏览.png图像文件的用户来说,已经提供了非常实用的功能。未来,若开发者计划拓展插件支持的文件格式或适用于其他浏览器,则需要考虑到对现有代码的扩展和兼容性测试。 总结: "ImagesOnGitHub-crx"插件通过创新的用户体验设计,解决了GitHub在浏览图像文件时的一些局限性,使得图像浏览更加直观和便捷。尽管目前该插件存在一些限制,如仅支持.png格式和仅在Chrome浏览器中可用,但它为用户和开发者提供了良好的思路和实践。对于希望提高效率和增强功能的用户来说,这类工具扩展了GitHub的实用性,是开发人员工具箱中的一个有益补充。
recommend-type

管理建模和仿真的文件

管理Boualem Benatallah引用此版本:布阿利姆·贝纳塔拉。管理建模和仿真。约瑟夫-傅立叶大学-格勒诺布尔第一大学,1996年。法语。NNT:电话:00345357HAL ID:电话:00345357https://theses.hal.science/tel-003453572008年12月9日提交HAL是一个多学科的开放存取档案馆,用于存放和传播科学研究论文,无论它们是否被公开。论文可以来自法国或国外的教学和研究机构,也可以来自公共或私人研究中心。L’archive ouverte pluridisciplinaire
recommend-type

【OPPO手机故障诊断专家】:工程指令快速定位与解决

![【OPPO手机故障诊断专家】:工程指令快速定位与解决](https://www.consumerelectronicstestdevelopment.com/media/2hlomnxy/oppo.jpg?anchor=center&mode=crop&width=1002&height=564&bgcolor=White&rnd=132773815380200000) # 摘要 本文综述了OPPO手机故障诊断的技术细节,涵盖了工程指令的基础理论、实践应用、高级技巧以及未来发展方向。首先介绍了工程指令的定义、分类、执行环境及其与手机系统交互的重要性。随后,深入探讨了工程指令在初步故障诊断
recommend-type

求[100,900]之间相差为12的素数对(注:要求素数对的两个素数均在该范围内)的个数

求解 [100, 900] 范围内相差为 12 的素数对,首先我们需要确定哪些数在这个区间内是素数。然后筛选出它们成对出现且差值为 12 的情况。 1. 确定素数范围内的素数:我们可以编写一个简单的程序来检查每个数字是否为素数,如果数字大于 1,并且除 2 到其平方根之间的所有整数都不能整除它,那么这个数字就是素数。 2. 遍历并寻找符合条件的素数对:从较大的素数开始向下遍历,找到的第一个素数作为“较大”素数,然后查看比它小 12 的下一个数,如果这个数也是素数,则找到了一对符合条件的素数。 3. 统计素数对的数量:统计在给定范围内找到的这种差距为 12 的素数对的数量。 由于计算素数
recommend-type

Android IPTV项目:直播频道的实时流媒体实现

资源摘要信息:"IPTV:直播IPTV的Android项目是一个基于Android平台的实时流式传输应用。该项目允许用户从M3U8或M3U格式的链接或文件中获取频道信息,并将这些频道以网格或列表的形式展示。用户可以在应用内选择并播放指定的频道。该项目的频道列表是从一个预设的列表中加载的,并且通过解析M3U或M3U8格式的文件来显示频道信息。开发者还计划未来更新中加入Exo播放器以及电子节目单功能,以增强用户体验。此项目使用了多种技术栈,包括Java、Kotlin以及Kotlin Android扩展。" 知识点详细说明: 1. IPTV技术: IPTV(Internet Protocol Television)即通过互联网协议提供的电视服务。它与传统的模拟或数字电视信号传输方式不同,IPTV通过互联网将电视内容以数据包的形式发送给用户。这种服务使得用户可以按需观看电视节目,包括直播频道、视频点播(VOD)、时移电视(Time-shifted TV)等。 2. Android开发: 该项目是针对Android平台的应用程序开发,涉及到使用Android SDK(软件开发工具包)进行应用设计和功能实现。Android应用开发通常使用Java或Kotlin语言,而本项目还特别使用了Kotlin Android扩展(Kotlin-Android)来优化开发流程。 3. 实时流式传输: 实时流式传输是指媒体内容以连续的流形式进行传输的技术。在IPTV应用中,实时流式传输保证了用户能够及时获得频道内容。该项目可能使用了HTTP、RTSP或其他流媒体协议来实现视频流的实时传输。 4. M3U/M3U8文件格式: M3U(Moving Picture Experts Group Audio Layer 3 Uniform Resource Locator)是一种常用于保存播放列表的文件格式。M3U8则是M3U格式的扩展版本,支持UTF-8编码,常用于苹果设备。在本项目中,M3U/M3U8文件被用来存储IPTV频道信息,如频道名称、视频流URL等。 5. Exo播放器: ExoPlayer是谷歌官方提供的一个开源视频播放器,专为Android优化。它支持多种特性,如自定义字幕、HDR视频播放、无缝直播等。ExoPlayer通常用于处理IPTV应用中的视频流媒体播放需求。 6. 电子节目单(EPG): 电子节目单是IPTV应用中一项重要功能,它为用户提供频道的节目指南,包括当前播放的节目以及未来节目的安排。电子节目单一般以网格或列表形式展示,方便用户浏览和搜索节目信息。 7. 开源贡献文化: 该项目提到了欢迎贡献者,表明这是一个开源项目。在开源文化中,开发者社区鼓励用户、开发者贡献代码来改进项目,这是一个共享知识、共同进步的过程。参与者通过贡献代码、报告问题或提供文档帮助等方式参与项目。 8. Kotlin编程语言: Kotlin是一种运行在Java虚拟机上的静态类型编程语言,它与Java完全兼容并可以无缝集成Java代码。Kotlin以其简洁、安全和富有表现力的特点被越来越多的Android开发者采用。在本项目中,使用Kotlin可以简化代码结构,提高开发效率和应用性能。 总结而言,本项目是一个面向Android平台的实时流媒体IPTV应用开发项目,它整合了实时流式传输、M3U/M3U8文件解析、Exo播放器使用、电子节目单功能等关键技术点,并在开源社区中寻求贡献者的参与。通过本项目,开发者可以深入了解如何在Android平台上实现IPTV服务,并学习到使用Kotlin和Java等编程语言进行Android应用开发的相关知识。
recommend-type

"互动学习:行动中的多样性与论文攻读经历"

多样性她- 事实上SCI NCES你的时间表ECOLEDO C Tora SC和NCESPOUR l’Ingén学习互动,互动学习以行动为中心的强化学习学会互动,互动学习,以行动为中心的强化学习计算机科学博士论文于2021年9月28日在Villeneuve d'Asq公开支持马修·瑟林评审团主席法布里斯·勒菲弗尔阿维尼翁大学教授论文指导奥利维尔·皮耶昆谷歌研究教授:智囊团论文联合主任菲利普·普雷教授,大学。里尔/CRISTAL/因里亚报告员奥利维耶·西格德索邦大学报告员卢多维奇·德诺耶教授,Facebook /索邦大学审查员越南圣迈IMT Atlantic高级讲师邀请弗洛里安·斯特鲁布博士,Deepmind对于那些及时看到自己错误的人...3谢谢你首先,我要感谢我的两位博士生导师Olivier和Philippe。奥利维尔,"站在巨人的肩膀上"这句话对你来说完全有意义了。从科学上讲,你知道在这篇论文的(许多)错误中,你是我可以依