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

Noise RETF Estimation and Removal for Low SNR Speech Enhancement 低信噪比语音增强的噪声RETF估计与去除

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596209

Lachlan I. Birnie, P. Samarasinghe, T. Abhayapala, Daniel Grixti-Cheng

引用次数: 1

Detecting Network State in the Presence of Varying Levels of Congestion 检测存在不同拥塞水平的网络状态

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596271

R. Fréin, Obinna Izima, Ali Malik

引用次数: 5

Multimodal Graph Coarsening for Interpretable, MRI-Based Brain Graph Neural Network 多模态图粗化可解释，基于核磁共振的脑图神经网络

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/MLSP52302.2021.9690626

Isaac Sebenius, Alexander Campbell, S. Morgan, E. Bullmore, Pietro Lio'

{"title":"Multimodal Graph Coarsening for Interpretable, MRI-Based Brain Graph Neural Network","authors":"Isaac Sebenius, Alexander Campbell, S. Morgan, E. Bullmore, Pietro Lio'","doi":"10.1109/MLSP52302.2021.9690626","DOIUrl":"https://doi.org/10.1109/MLSP52302.2021.9690626","url":null,"abstract":"Graph neural networks (GNN s) are a powerful class of model for representation learning on relational data and graph-structured signal, such as brain connectivity graphs derived from neuroimaging. To date, existing work applying graph learning methods to brain connectivity is limited to a single neuroimaging modality such as structural or functional MRI. In practice, the brain is best represented by multiple networks arising from different imaging modalities. We develop a gen-eral framework for jointly pooling multimodal graphs which share the same set of underlying nodes whilst differing in edge connectivity. Building on this approach, we propose a multimodal GNN (MM-GNN) model that incorporates mul-tiple types of neuroimaging-based brain connectivity. When applied to the task of classifying brain images from patients with schizophrenia and healthy control subjects, we observe that incorporating multimodal pooling dramatically improves performance over non-pooled networks and that MM-GNN matches or improves performance over multiple single-modal and non-GNN baselines. Finally, we demonstrate how our approach uses multimodal data to learn a unified, interpretable measure of the salience of individual brain regions of interest. In this way, MM-GNN represents a new method for leveraging diverse brain connectivity data to enhance the detection of mental health disorders and to understand their biological underpinnings.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"81 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":"116134939","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

A Weighted Loss Function to Predict Control Parameters for Supercontinuum Generation Via Neural Networks 基于加权损失函数的神经网络超连续统生成控制参数预测

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596142

Diego Stucchi, Andrea Corsini, G. Genty, G. Boracchi, A. Foi

{"title":"A Weighted Loss Function to Predict Control Parameters for Supercontinuum Generation Via Neural Networks","authors":"Diego Stucchi, Andrea Corsini, G. Genty, G. Boracchi, A. Foi","doi":"10.1109/mlsp52302.2021.9596142","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596142","url":null,"abstract":"Supercontinuum light is generated by a train of laser pulses propagating in an optical fiber. The parameters characterizing these pulses influence the spectrum of the light as it exits the fiber. While spectrum generation is a direct process governed by nonlinear equations that can be reproduced through numerical simulation, determining the parameters of the pulse generating a given spectrum is a difficult inverse problem. Solving this inverse problem has a relevant practical implication, as it allows generating beams with desired spectral properties. We solve this multidimensional parameter estimation problem by training a neural network and we introduce, as key technical contribution, a weighted loss function that improves the estimation accuracy. Most remarkably, this loss function is not specific to the considered supercontinuum scenario, but has the potential to improve solutions of similar inverse problems where the forward process can be reproduced via computationally demanding simulations. Our experiments demonstrate the effectiveness of the pursued approach and of our weighted loss function.","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":"129677460","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

Scalable Community Detection in the Degree-Corrected Stochastic Block Model 度校正随机块模型中的可扩展社区检测

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596377

Yicong He, Andre Beckus, George K. Atia

{"title":"Scalable Community Detection in the Degree-Corrected Stochastic Block Model","authors":"Yicong He, Andre Beckus, George K. Atia","doi":"10.1109/mlsp52302.2021.9596377","DOIUrl":"https://doi.org/10.1109/mlsp52302.2021.9596377","url":null,"abstract":"Community detection aims to partition a connected graph into a small number of clusters. The Degree-Corrected Stochastic Block Model (DCSBM) is one popular generative model that yields graphs with varying degree distributions within the communities. However, large computational complexity and storage requirements of existing approaches for DCSBM limit their scalability to large graphs. In this paper, we advance a scalable framework for DCSBM, in which the full graph is first sub-sampled by selecting a small subset of the nodes, then a clustering of the induced subgraph is obtained, followed by low-complexity retrieval of the global community structure from the clustering of the graph sketch. To sample the underlying graph, we introduce a family of sampling schemes that capture local community structures using metrics derived from the average neighbor degrees, which are shown to achieve the twin objective of sampling from low-density clusters and identifying high-degree nodes within each cluster. The proposed approach can perform on par with full scale clustering while affording substantial complexity and storage gains as demonstrated through experiments using both synthetic and real data.","PeriodicalId":156116,"journal":{"name":"2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP)","volume":"120 3","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2021-10-25","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"114001613","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

Deep Complex Convolutional Recurrent Network for Multi-Channel Speech Enhancement and Dereverberation 用于多通道语音增强和去噪的深度复杂卷积循环网络

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596086

Femke B. Gelderblom, T. A. Myrvoll

引用次数: 0

A Coarse-to-Fine Object Detection Framework for High-Resolution Images with Sparse Objects 基于稀疏目标的高分辨率图像粗到细目标检测框架

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596518

Jinyan Liu, Longbin Yan, Jie Chen

引用次数: 0

Optimized Spatial Matching for Visual Object Tracking 视觉目标跟踪的优化空间匹配

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596182

Fuxiang Wang, Qing Mei, Xuhui Liu, Yao Xiao

引用次数: 0

Bayesian Two-Stage Sequential Change Diagnosis Via Multi-Sensor Array 基于多传感器阵列的贝叶斯两阶段序列变化诊断

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596446

Xiaochuan Ma, L. Lai, Shuguang Cui

引用次数: 2

Small Moving Target MOT Tracking with GM-PHD Filter and Attention-Based CNN 基于GM-PHD滤波和基于注意力的CNN的小运动目标MOT跟踪

2021 IEEE 31st International Workshop on Machine Learning for Signal Processing (MLSP) Pub Date : 2021-10-25 DOI: 10.1109/mlsp52302.2021.9596204

Camilo Aguilar, M. Ortner, J. Zerubia

引用次数: 6