2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)最新文献_第4页

Real-Time Optimizing Weighted Gaussian Curvature for 4K Videos 4K视频实时优化加权高斯曲率

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596473

Wenming Tang, Lebin Zhou, Yuanhao Gong

引用次数: 8

Introducing K-Anonymity Principles to Adversarial Attacks for Privacy Protection in Image Classification Problems 将k -匿名原理引入图像分类问题中隐私保护的对抗性攻击

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596565

V. Mygdalis, A. Tefas, I. Pitas

{"title":"Introducing K-Anonymity Principles to Adversarial Attacks for Privacy Protection in Image Classification Problems","authors":"V. Mygdalis, A. Tefas, I. Pitas","doi":"10.1109/mlsp52302.2021.9596565","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596565","url":null,"abstract":"The network output activation values for a given input can be employed to produce a sorted ranking. Adversarial attacks typically generate the least amount of perturbation required to change the classifier label. In that sense, generated adversarial attack perturbation only affects the output in the 1st sorted ranking position. We argue that meaningful information about the adversarial examples i.e., their original labels, is still encoded in the network output ranking and could potentially be extracted, using rule-based reasoning. To this end, we introduce a novel adversarial attack methodology inspired by the K-anonymity principles, that generates adversarial examples that are not only misclassified, but their output sorted ranking spreads uniformly along K different positions. Any additional perturbation arising from the strength of the proposed objectives, is regularized by a visual similarity-based term. Experimental results denote that the proposed approach achieves the optimization goals inspired by K-anonymity with reduced perturbation as well.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"92 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":"132104106","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: 3

Optimizing Time Domain Fully Convolutional Networks for 3D Speech Enhancement in a Reverberant Environment Using Perceptual Losses 利用感知损失优化时域全卷积网络在混响环境下的三维语音增强

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596103

Heitor R. Guimarães, Wesley Beccaro, M. A. Ramírez

引用次数: 3

Bag of Groups of Convolutional Features Model for Visual Object Recognition 用于视觉目标识别的卷积特征包模型

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596432

Jaspreet Singh, Chandan Singh

引用次数: 1

Zero-Shot Motion Pattern Recognition from 4D Point-Clouds 4D点云的零射击运动模式识别

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596375

Dariush Salami, S. Sigg

引用次数: 2

Compression of DNNs Using Magnitude Pruning and Nonlinear Information Bottleneck Training 基于幅度修剪和非线性信息瓶颈训练的深度神经网络压缩

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596128

Morten Østergaard Nielsen, Jan Østergaard, J. Jensen, Z. Tan

{"title":"Compression of DNNs Using Magnitude Pruning and Nonlinear Information Bottleneck Training","authors":"Morten Østergaard Nielsen, Jan Østergaard, J. Jensen, Z. Tan","doi":"10.1109/mlsp52302.2021.9596128","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596128","url":null,"abstract":"As Deep Neural Networks (DNNs) have achieved state-of-the-art performance in various scientific fields and applications, the memory and computational complexity of DNNs have increased concurrently. The increased complexity required by DNNs prohibits them from running on platforms with limited computational resources. This has sparked a renewed interest in parameter pruning. We propose to replace the standard cross-entropy objective - typically used in classification problems - with the Nonlinear Information Bottleneck (NIB) objective to improve the accuracy of a pruned network. We demonstrate, that our proposal outperforms cross-entropy combined with global magnitude pruning for high compression rates on VGG-nets trained on CIFAR10. With approximately 97% of the parameters pruned, we obtain an accuracy of 87.63% and 88.22% for VGG-16 and VGG-19, respectively, where the baseline accuracy is 91.5% for the unpruned networks. We observe that the majority of biases are pruned completely, and pruning parameters globally outperforms layer-wise pruning.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"69 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":"116953041","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

About the Equivalence Between Complex-Valued and Real-Valued Fully Connected Neural Networks - Application to Polinsar Images 关于复值与实值全连接神经网络的等价性——在Polinsar图像中的应用

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596542

J. A. Barrachina, C. Ren, G. Vieillard, C. Morisseau, J. Ovarlez

引用次数: 7

Distributed Dictionary Learning Over Heterogeneous Clients Using Local Adaptive Dictionaries 使用本地自适应字典的异构客户端分布式字典学习

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596408

You-De Huang, Yao-Win Peter Hong

{"title":"Distributed Dictionary Learning Over Heterogeneous Clients Using Local Adaptive Dictionaries","authors":"You-De Huang, Yao-Win Peter Hong","doi":"10.1109/mlsp52302.2021.9596408","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596408","url":null,"abstract":"This work examines the use of dictionary learning among distributed clients with heterogeneous tasks. We propose a distributed dictionary learning algorithm that enables collaborative training of a shared global dictionary among clients while adaptively constructing local dictionary elements to address the heterogeneity of local tasks. The proposed distributed dictionary learning with local adaptive dictionaries (DDL-LAD) algorithm consists of two parts: a distributed optimization procedure that enables joint training of the dictionaries without sharing of the local datasets with the server, and a splitting and elimination procedure that is used to adaptively construct local dictionary elements. The splitting procedure identifies elements in the global dictionary that exhibit discriminative features for the local tasks. The elements are split and appended to the local dictionaries. Then, to avoid overgrowing of the local dictionaries, an elimination procedure is adopted to prune elements with less usage. Experiments on a distributed EMNIST dataset is provided to demonstrate the effectiveness of the proposed DDL-LAD algorithm compared to existing schemes that adopt only a global shared dictionary.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"1 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":"129055462","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: 0

Toward Minimal-Sufficiency in Regression Tasks: An Approach Based on a Variational Estimation Bottleneck 回归任务的最小充分性:基于变分估计瓶颈的方法

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596368

Zhaoyang Lyu, Gholamali Aminian, M. Rodrigues

引用次数: 0

Hyphylearn: A Domain Adaptation-Inspired Approach to Classification Using Limited Number of Training Samples Hyphylearn:一种基于领域自适应的有限训练样本分类方法

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596469

Alireza Nooraiepour, W. Bajwa, N. Mandayam

{"title":"Hyphylearn: A Domain Adaptation-Inspired Approach to Classification Using Limited Number of Training Samples","authors":"Alireza Nooraiepour, W. Bajwa, N. Mandayam","doi":"10.1109/mlsp52302.2021.9596469","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596469","url":null,"abstract":"The fundamental task of classification given a limited number of training data samples is considered for physical systems with known parametric statistical models. As a solution, a hybrid classification method-termed HYPHYLEARN-is proposed that exploits both the physics-based statistical models and the learning-based classifiers. The proposed solution is based on the conjecture that HYPHYLEARN would alleviate the challenges associated with the individual approaches of learning-based and statistical model-based classifiers by fusing their respective strengths. The proposed hybrid approach first estimates the unobservable model parameters using the available (suboptimal) statistical estimation procedures, and subsequently uses the physics-based statistical models to generate synthetic data. Next, the training data samples are incorporated with the synthetic data in a learning-based classifier that is based on domain-adversarial training of neural networks. Numerical results on multiuser detection, a concrete communication problem, demonstrate that HYPHYLEARN leads to major classification improvements compared to the existing stand-alone and hybrid classification methods.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"45 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":"121561557","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