2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)最新文献_第4页

A Simple Gaussian Kernel Classifier with Automated Hyperparameter Tuning 具有自动超参数调优的简单高斯核分类器

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023147

K. Fukumori, Toshihisa Tanaka

{"title":"A Simple Gaussian Kernel Classifier with Automated Hyperparameter Tuning","authors":"K. Fukumori, Toshihisa Tanaka","doi":"10.1109/APSIPAASC47483.2019.9023147","DOIUrl":"https://doi.org/10.1109/APSIPAASC47483.2019.9023147","url":null,"abstract":"This paper establishes a fitting method for a kernel logistic regression model that uses generalized Gaussian kernel and its parameter optimization method. Kernel logistic regression is a classification model that uses kernel methods effectively. This is one of the methods to construct an effective nonlinear system with a reproducing kernel Hilbert space (RKHS) induced from positive semi-definite kernels. Most classifiers that are combined with Gaussian kernel functions generally assume uncorrelatedness within the feature vectors. Thus, the Gaussian kernel consists of only two parameters (namely, mean and precision). In this paper, we propose a model using a generalized Gaussian kernel represented flexibly in each dimension of feature vector. In addition, the parameters of kernel are fully data-driven. For the fitting of proposed model, an ℓ1-regularization is introduced to supress the number of support vectors. A numerical experiment showed that the classification performance of the proposed model is almost the same as RBF-SVM even though the proposed model has a small number of support vectors.","PeriodicalId":145222,"journal":{"name":"2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","volume":"26 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125803747","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

SHNU Anti-spoofing Systems for ASVspoof 2019 Challenge 面向asvspoof2019挑战赛的SHNU抗欺骗系统

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023319

Zhimin Feng, Qiqi Tong, Yanhua Long, Shuang Wei, Chunxia Yang, Qiaozheng Zhang

{"title":"SHNU Anti-spoofing Systems for ASVspoof 2019 Challenge","authors":"Zhimin Feng, Qiqi Tong, Yanhua Long, Shuang Wei, Chunxia Yang, Qiaozheng Zhang","doi":"10.1109/APSIPAASC47483.2019.9023319","DOIUrl":"https://doi.org/10.1109/APSIPAASC47483.2019.9023319","url":null,"abstract":"This paper presents an experimental analysis of SHNU anti-spoofing systems for the ASVspoof 2019 challenge. This challenge focused on countermeasures for three major attack types, namely those stemming from the advanced technology of TTS, VC and replay spoofing attacks. According to the type of attacks, the challenge was divided into two independent sub-challenges, the logical access (LA) and physical access (PA). Results of different anti-spoofing technologies on both sub-challenges were reported. Furthermore, the same countermeasures were also evaluated on two previous challenges, the ASVspoof 2015 and 2017. Experiments on cross-databases showed that, it appeared hard to generalize the classifiers trained from ASVspoof 2019 LA and PA databases to the previous challenges. The generalization ability of anti-spoofing technologies to different, new and unknown conditions was still very challenging. In addition, the effectiveness of different acoustic features were also examined and reported. Finally, we investigated the linear and an interfusing score-level fusion methods to individual systems to achieve better performance.","PeriodicalId":145222,"journal":{"name":"2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","volume":"70 5 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125977619","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

Differentiable Programming based Step Size Optimization for LMS and NLMS Algorithms 基于可微规划的LMS和NLMS算法步长优化

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023175

K. Hayashi, Kaede Shiohara, Tetsuya Sasaki

{"title":"Differentiable Programming based Step Size Optimization for LMS and NLMS Algorithms","authors":"K. Hayashi, Kaede Shiohara, Tetsuya Sasaki","doi":"10.1109/APSIPAASC47483.2019.9023175","DOIUrl":"https://doi.org/10.1109/APSIPAASC47483.2019.9023175","url":null,"abstract":"We propose TLMS (Trainable Least Mean Squares) and TNLMS (Trainable Normalized LMS) algorithms, which use different step size parameter at each iteration determined by machine learning approach. It has been known that LMS algorithm can achieve fast convergence and small steady-state error simultaneously by dynamically controlling the step size compared as a fix step size, however, in conventional variable step size approaches, the step size parameter has been controlled in rather heuristic manners. In this study, based on the concept of differential programming, we unfold the iterative process of LMS or NLMS algorithms, and obtain a multilayer signal-flow graph similar to a neural network, where each layer has a step size of each iteration of LMS or NLMS algorithm as an independent learnable parameter. Then, we optimize the step size parameters of all iterations by using a machine learning approach, such as the stochastic gradient descent. Numerical experiments demonstrate the performance of the proposed TLMS and TNLMS algorithms under various conditions.","PeriodicalId":145222,"journal":{"name":"2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","volume":"29 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"126138805","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

A Novel Effective Dimensionality Reduction Algorithm for Water Chiller Fault Data 一种新的冷水机组故障数据降维算法

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023322

Zhuozheng Wang, Yingjie Dong, Wei Liu

引用次数: 1

GSC Based Speech Enhancement with Generative Adversarial Network 基于GSC的生成对抗网络语音增强

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023115

Yaofeng Zhou, C. Bao, Rui Cheng

引用次数: 0

Classification of Polarimetric SAR Image based on Improved Fuzzy Clustering 基于改进模糊聚类的极化SAR图像分类

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023152

Zheng Cheng, Ping Han, Binbin Han, Jiahui Sun

引用次数: 0

Prosodic Cues in the Interpretation of Echo Questions in Chinese Spoken Dialogues 汉语口语对话中回声疑问句翻译的韵律线索

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023150

Ai-jun Li, G. Huang, Zhiqiang Li

{"title":"Prosodic Cues in the Interpretation of Echo Questions in Chinese Spoken Dialogues","authors":"Ai-jun Li, G. Huang, Zhiqiang Li","doi":"10.1109/APSIPAASC47483.2019.9023150","DOIUrl":"https://doi.org/10.1109/APSIPAASC47483.2019.9023150","url":null,"abstract":"This study examines the effect of prosodic cues in the disambiguation of five discourse-pragmatic functions of echo questions and the corresponding statements in Chinese spoken dialogues. Data were collected in a “role-play” format to mimic different communicative functions of echo questions in real-life situations. Statistical analyses were performed on both global and local F0 variations associated with intonation patterns in echo questions and corresponding statements. Results showed that boundary tone features alone are not good predictors in distinguishing echo questions and statements; variations in intonation patterns are related to the different discourse-pragmatic functions that echo questions serve; echo questions and statements, as well as different discourse-pragmatic functions of echo questions, can be distinguished on the basis of global variations of prosodic features such as overall F0 slope and average F0, combined with local changes due to boundary tone features; and when information about morpho-syntactic structures and boundary tone features were included in the analysis, the accuracy of discriminant analysis was at 76.5%∼94.1% for statements and echo questions, and at 57.6%∼83.5% for different discourse-pragmatic functions. The accuracy dropped to 70.9% (2 groups) and 40.9% (6 groups) when morpho-syntactic structural information was not included, indicating that structural and contextual information contributed 30% and 60% respectively.","PeriodicalId":145222,"journal":{"name":"2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","volume":"73 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"132489646","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

A LSTM-Based Joint Progressive Learning Framework for Simultaneous Speech Dereverberation and Denoising 基于lstm的同步语音去噪联合渐进学习框架

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023160

Xin Tang, Jun Du, Li Chai, Yannan Wang, Qing Wang, Chin-Hui Lee

{"title":"A LSTM-Based Joint Progressive Learning Framework for Simultaneous Speech Dereverberation and Denoising","authors":"Xin Tang, Jun Du, Li Chai, Yannan Wang, Qing Wang, Chin-Hui Lee","doi":"10.1109/APSIPAASC47483.2019.9023160","DOIUrl":"https://doi.org/10.1109/APSIPAASC47483.2019.9023160","url":null,"abstract":"We propose a joint progressive learning (JPL) framework of gradually mapping highly noisy and reverberant speech features to less noisy and less reverberant speech features in a layer-by-layer stacking scenario for simultaneous speech denoising and dereverberation. As such layers are easier to learn than mapping highly distorted speech features directly to clean and anechoic speech features, we adopt a divide-and-conquer learning strategy based on a long short-term memory (LSTM) architecture, and explicitly design multiple intermediate target layers. Each hidden layer of the LSTM network is guided by a step-by-step signal-to-noise-ratio (SNR) increase and reverberant time decrease. Moreover, post-processing is applied to further improve the enhancement performance by averaging the estimated intermediate targets. Experiments demonstrate that the proposed JPL approach not only improves objective measures for speech quality and intelligibility, but also achieves a more compact model design when compared to the direct mapping and two-stage, namely denoising followed dereverberation approaches.","PeriodicalId":145222,"journal":{"name":"2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","volume":"21 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"129288640","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

Acquisition of L2 Mandarin Rhythm By Russian and Japanese Learners 俄语和日语学习者汉语第二语言节奏的习得

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023198

Yiran Ding, Yanlu Xie, Jinsong Zhang

引用次数: 0

Statistical analysis on characteristic whisker movements observed in reward processing 奖励加工中特征须运动的统计分析

2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC) Pub Date : 2019-11-01 DOI: 10.1109/APSIPAASC47483.2019.9023135

J. Yoshimoto, Jumpei Ozaki, Kohta Mizutani, Takashi Nakano, K. Ikeda, Takayuki Yamashita

{"title":"Statistical analysis on characteristic whisker movements observed in reward processing","authors":"J. Yoshimoto, Jumpei Ozaki, Kohta Mizutani, Takashi Nakano, K. Ikeda, Takayuki Yamashita","doi":"10.1109/APSIPAASC47483.2019.9023135","DOIUrl":"https://doi.org/10.1109/APSIPAASC47483.2019.9023135","url":null,"abstract":"Internal states of the brain can be often reflected as facial expressions. However, how animals show their facial expression is largely unexplored. Here, we focus on mice and investigate whether their whisker movements could be a facial expression of their internal states related to reward processing. We trained three mice for an auditory association task and filmed their whiskers during the task performance after enough learning. We found that approximately 5–8 Hz periodic whisking was commonly observed during reward-associated Go cue presentation. Such whisking rarely occurred in No-Go cue trials or in Go cue trials where the mice were not motivated to get a reward. Furthermore, after acquiring a reward, the mice whisked with a more protracted set-point. Using machine learning, we could accurately indicate reward-anticipating and reward-acquiring trials only from whisker time plots. Our analyses suggest that mice exhibit stereotypic whisker movements as a part of orofacial movements related to reward anticipation and acquisition.","PeriodicalId":145222,"journal":{"name":"2019 Asia-Pacific Signal and Information Processing Association Annual Summit and Conference (APSIPA ASC)","volume":"118 1","pages":"0"},"PeriodicalIF":0.0,"publicationDate":"2019-11-01","publicationTypes":"Journal Article","fieldsOfStudy":null,"isOpenAccess":false,"openAccessPdf":"","citationCount":null,"resultStr":null,"platform":"Semanticscholar","paperid":"125557819","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