2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)最新文献_第8页

Distributed Variational Information Bottleneck for IOT Environments 物联网环境中的分布式变分信息瓶颈

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596553

Zahir Alsulaimawi, Huaping Liu

{"title":"Distributed Variational Information Bottleneck for IOT Environments","authors":"Zahir Alsulaimawi, Huaping Liu","doi":"10.1109/mlsp52302.2021.9596553","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596553","url":null,"abstract":"Deep learning is becoming the latest trend in sensitive applications, such as healthcare, criminal justice, and finance. As these new applications emerge, adversaries are developing ways to circumvent them. In this paper, we investigate users' revealing data to the public; parts of it are often sensitive when compactly represented. The representation should ensure that the target information is served accurately and reliably while simultaneously safeguarding sensitive information. In order to achieve that goal, we present a supervised deep learning framework based on the Information Bottleneck (IB) principle. The purpose of this was to maximize the mutual information between utility labels, and the learned compressing representation while minimizing the mutual information between the learned compressing representation and the original representation. Additionally, we examine a distributed learning framework to securely aggregate data from the Internet of Things (IoT) devices and create a utility model that is compatible with IoT devices. We apply the variational mutual information approximation to gain an accurate representation of bottlenecks. Through experiments with synthetic datasets, we demonstrate the efficiency and privacy-preserving capabilities of our framework.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"36 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"133138602","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Random Feature Approximation for Online Nonlinear Graph Topology Identification 在线非线性图拓扑识别的随机特征逼近

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-19 DOI: 10.1109/mlsp52302.2021.9596512

Rohan Money, Joshin P. Krishnan, B. Beferull-Lozano

引用次数: 7

Not All Noise is Accounted Equally: How Differentially Private Learning Benefits from Large Sampling Rates 并非所有的噪音都是平等的:私人学习如何从大采样率中获益

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-12 DOI: 10.1109/mlsp52302.2021.9596307

Friedrich Dörmann, Osvald Frisk, L. Andersen, Christian Fischer Pedersen

{"title":"Not All Noise is Accounted Equally: How Differentially Private Learning Benefits from Large Sampling Rates","authors":"Friedrich Dörmann, Osvald Frisk, L. Andersen, Christian Fischer Pedersen","doi":"10.1109/mlsp52302.2021.9596307","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596307","url":null,"abstract":"Learning often involves sensitive data and as such, privacy preserving extensions to Stochastic Gradient Descent (SGD) and other machine learning algorithms have been developed using the definitions of Differential Privacy (DP). In differentially private SGD, the gradients computed at each training iteration are subject to two different types of noise. Firstly, inherent sampling noise arising from the use of minibatches. Secondly, additive Gaussian noise from the underlying mechanisms that introduce privacy. In this study, we show that these two types of noise are equivalent in their effect on the utility of private neural networks, however they are not accounted for equally in the privacy budget. Given this observation, we propose a training paradigm that shifts the proportions of noise towards less inherent and more additive noise, such that more of the overall noise can be accounted for in the privacy budget. With this paradigm, we are able to improve on the state-of-the-art in the privacy/utility tradeoff of private end-to-end CNNs.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"33 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-12","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"127748083","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 14

Joint Optimization of System Design and Reconstruction in MIMO Radar Imaging MIMO雷达成像系统设计与重构的联合优化

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-07 DOI: 10.1109/mlsp52302.2021.9596168

Tomer Weiss, Nissim Peretz, S. Vedula, A. Feuer, A. Bronstein

{"title":"Joint Optimization of System Design and Reconstruction in MIMO Radar Imaging","authors":"Tomer Weiss, Nissim Peretz, S. Vedula, A. Feuer, A. Bronstein","doi":"10.1109/mlsp52302.2021.9596168","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596168","url":null,"abstract":"Multiple-input multiple-output (MIMO) radar is one of the leading depth sensing modalities. However, the usage of multiple receive channels lead to relative high costs and prevent the penetration of MIMOs in many areas such as the automotive industry. Over the last years, few studies concentrated on designing reduced measurement schemes and image reconstruction schemes for MIMO radars, however these problems have been so far addressed separately. On the other hand, recent works in optical computational imaging have demonstrated growing success of simultaneous learning-based design of the acquisition and reconstruction schemes, manifesting significant improvement in the reconstruction quality. Inspired by these successes, in this work, we propose to learn MIMO acquisition parameters in the form of receive (Rx) antenna elements locations jointly with an image neural-network based reconstruction. To this end, we propose an algorithm for training the combined acquisition-reconstruction pipeline end-to-end in a differentiable way. We demonstrate the significance of using our learned acquisition parameters with and without the neural-network reconstruction. Code and datasets will be released upon publication.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"74 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122282427","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 2

Bi-Rads-Net: An Explainable Multitask Learning Approach for Cancer Diagnosis in Breast Ultrasound Images Bi-Rads-Net:一种可解释的多任务学习方法用于乳腺超声图像的癌症诊断

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-01 DOI: 10.1109/mlsp52302.2021.9596314

Boyu Zhang, Aleksandar Vakanski, Min Xian

引用次数: 4

Adaptive Approach for Sparse Representations Using the Locally Competitive Algorithm for Audio 基于局部竞争算法的音频稀疏表示自适应方法

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-09-29 DOI: 10.1109/MLSP52302.2021.9596348

Soufiyan Bahadi, J. Rouat, É. Plourde

引用次数: 0

Learning-Based UE Classification in Millimeter-Wave Cellular Systems with Mobility 基于学习的移动毫米波蜂窝系统UE分类

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-09-13 DOI: 10.1109/mlsp52302.2021.9596275

Dino Pjani'c, Alexandros Sopasakis, H. Tataria, F. Tufvesson, A. Reial

引用次数: 0

Differentially Private Variable Selection via the Knockoff Filter 通过仿冒过滤器进行差异私有变量选择

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-09-12 DOI: 10.1109/mlsp52302.2021.9596470

Mehrdad Pournaderi, Yu Xiang

引用次数: 2

Few-Shot Learning Via Dependency Maximization and Instance Discriminant Analysis 基于依赖最大化和实例判别分析的少射学习

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-09-07 DOI: 10.1109/mlsp52302.2021.9596284

Zejiang Hou, S. Kung

{"title":"Few-Shot Learning Via Dependency Maximization and Instance Discriminant Analysis","authors":"Zejiang Hou, S. Kung","doi":"10.1109/mlsp52302.2021.9596284","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596284","url":null,"abstract":"We study the few-shot learning (FSL) problem, where a model learns to recognize new objects with extremely few labeled training data per category. Most of previous FSL approaches resort to the meta-learning paradigm, where the model accumulates inductive bias through learning many training tasks so as to solve a new unseen few-shot task. In contrast, we propose a simple approach to exploit unlabeled data accompanying the few-shot task for improving few-shot performance. Firstly, we propose a Dependency Maximization method based on the Hilbert-Schmidt norm of the cross-covariance operator, which maximizes the statistical dependency between the embedded feature of those unlabeled data and their label predictions, together with the supervised loss over the support set. We then use the obtained model to infer the pseudo-labels for those unlabeled data. Furthermore, we propose an Instance Discriminant Analysis to evaluate the credibility of each pseudo-labeled example and select the most faithful ones into an augmented support set to retrain the model as in the first step. We iterate the above process until the pseudo-labels for the unlabeled data become stable. Following the standard transductive and semi-supervised FSL setting, our experiments show that the proposed method outperforms previous state-of-the-art methods on four widely used benchmarks, including mini-ImageNet, tiered-ImageNet, CUB, and CIFARFS.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"11 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-09-07","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"122034657","PeriodicalName":null,"FirstCategoryId":null,"ListUrlMain":null,"RegionNum":0,"RegionCategory":"","ArticlePicture":[],"TitleCN":null,"AbstractTextCN":null,"PMCID":"","EPubDate":null,"PubModel":null,"JCR":null,"JCRName":null,"Score":null,"Total":0}

引用次数: 1

Self-Attention for Audio Super-Resolution 音频超分辨率的自我关注

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-08-26 DOI: 10.1109/mlsp52302.2021.9596082

Nathanaël Carraz Rakotonirina

引用次数: 10