Robustness Analysis of Deep Learning Frameworks on Mobile Platforms
Department of Electrical and Software Engineering, University of Calgary, Canada(卡尔加里大学电气与软件工程系,加拿大卡尔加里大学(注:此处按要求提取到大学级别,“Department of Electrical and Software Engineering”是系,但整体机构为University of Calgary,故保留原机构名))
With the recent increase in the computational power of modern mobile devices, machine learning-based heavy tasks such as face detection and speech recognition are now integral parts of such devices. This requires frameworks to execute machine learning models (e.g., Deep Neural Networks) on mobile devices. Although there exist studies on the accuracy and performance of these frameworks, the quality of on-device deep learning frameworks, in terms of their robustness, has not been systematically studied yet. In this paper, we empirically compare two on-device deep learning frameworks with three adversarial attacks on three different model architectures. We also use both the quantized and unquantized variants for each architecture. The results show that, in general, neither of the deep learning frameworks is better than the other in terms of robustness, and there is not a significant difference between the PC and mobile frameworks either. However, in cases like Boundary attack, mobile version is more robust than PC. In addition, quantization improves robustness in all cases when moving from PC to mobile.
SeesawFaceNets: sparse and robust face verification model for mobile platform
DiDiChuxing(滴滴出行)
深度卷积神经网络(DCNN)是计算机视觉相关任务中使用最广泛的解决方案,其中最重要的应用场景之一是人脸验证。由于其高精度的性能,深度人脸验证模型的推理阶段发生在云平台上通过互联网在大多数实际场景中起着关键作用。然而,存在两个关键问题:首先,个人隐私可能没有得到很好的保护,因为他们必须将个人照片和其他私人信息上传到在线云后端。其次,无论是训练还是推断阶段都是耗时的,并且延迟可能会影响客户体验,特别是当互联网链路速度不太稳定或在移动接收不太好的偏远地区时,而且在建筑物和其他建筑物可能阻挡移动信号的城市中。因此,设计具有低内存需求和计算成本的轻量级网络是移动平台上人脸验证最实用的解决方案之一。在本文中,一种新的移动网络名为SeesawFaceNets,一个简单而有效的模型,提出了高效地部署移动设备的人脸识别。密集的实验结果表明,我们提出的模型SeesawFaceNets优于基线MobileFaceNets,只有{ \bf 66\%}(146 M VS 221 M MAds)计算成本,更小的批量大小和更少的训练步骤,和SeesawFaceNets实现了与其他SOTA模型相当的性能,例如仅具有{ \bf 54.2\%}的mobiface(1.3 M VS 2.4 M)参数和{ \bf 31.6\%}(146 M VS 462 M MAdds)计算成本,它最终也在所有5个上市的公共验证数据集上…
A Semi-Supervised and Inductive Embedding Model for Churn Prediction of Large-Scale Mobile Games
Texas A&M University(德州农工大学);Samsung Research America(三星美国研究院);University of Pittsburgh(匹兹堡大学);The University of Arizona(亚利桑那大学)
移动游戏已经成为一个有前途的市场,收入达数十亿美元。世界各地开发了各种移动游戏平台和服务。这些平台和服务面临的一个关键挑战是了解手机游戏中的用户流失行为。准确的流失预测将使许多利益相关者受益,如游戏开发商,广告商和平台运营商。在本文中,我们提出了第一个大规模的流失预测解决方案的手机游戏。鉴于建立在传统机器学习模型基础上的最先进方法的共同局限性,我们设计了一种新的半监督和归纳嵌入模型,该模型联合学习用户-应用关系的预测函数和嵌入函数。我们通过具有独特边缘嵌入技术的深度神经网络对这两个功能进行建模,该技术能够捕获上下文信息和关系动态。我们还设计了一个新的属性随机游走技术,同时考虑到拓扑邻接和属性相似性。为了评估我们的解决方案的性能,我们从Samsung Game Launcher平台收集了真实数据,其中包括数万款游戏和数亿次用户与应用程序的交互。这些数据的实验结果表明,我们提出的模型对现有的最先进的方法的优越性。
TRAMBA: A Hybrid Transformer and Mamba Architecture for Practical Audio and Bone Conduction Speech Super Resolution and Enhancement on Mobile and Wearable Platforms
MatchNAS: Optimizing Edge AI in Sparse-Label Data Contexts via Automating Deep Neural Network Porting for Mobile Deployment
School of Computer Science and Engineering, University of New South Wales(新南威尔士大学计算机科学与工程学院);Faculty of Engineering & Information Technology, University of Technology Sydney(悉尼科技大学工程与信息技术学院);CSIRO’s Data61(澳大利亚联邦科学与工业研究组织Data61);University of New South Wales(新南威尔士大学)
Catch the Ball: Accurate High-Speed Motions for Mobile Manipulators via Inverse Dynamics Learning
University of Toronto Institute for Aerospace Studies (UTIAS)(多伦多大学航空航天研究所(UTIAS));Vector Institute for artificial Intelligence(人工智能向量研究所);University of Toronto(多伦多大学)
A Cloud-based Deep Learning Framework for Remote Detection of Diabetic Foot Ulcers
Manchester Metropolitan University(曼彻斯特城市大学);Lancashire Teaching Hospitals NHS Foundation Trust(兰开夏教学医院NHS基金会信托);The University of Manchester(曼彻斯特大学);Manchester University NHS Foundation Trust(曼彻斯特大学NHS基金会信托);Salford Royal NHS Foundation Trust(索尔福德皇家NHS基金会信托)
Reconfigurable Intelligent Surface Assisted Mobile Edge Computing with Heterogeneous Learning Tasks
Southern University of Science and Technology(南方科技大学);The University of Hong Kong(香港大学);Peng Cheng Laboratory(鹏城实验室);South China University of Technology(华南理工大学)