scholarly journals N-MTTL SI Model: Non-Intrusive Multi-Task Transfer Learning-Based Speech Intelligibility Prediction Model with Scenery Classification

2021 ◽  
Author(s):  
Ĺuboš Marcinek ◽  
Michael Stone ◽  
Rebecca Millman ◽  
Patrick Gaydecki
Keyword(s):  
Prediction Model ◽  
Transfer Learning ◽  
Speech Intelligibility ◽  
Task Transfer ◽  
Si Model

A multi-task transfer learning method with dictionary learning

2020 ◽  
Vol 191 ◽  
pp. 105233 ◽  
Author(s):  
Xin Zheng ◽  
Luyue Lin ◽  
Bo Liu ◽  
Yanshan Xiao ◽  
Xiaoming Xiong
Keyword(s):  
Transfer Learning ◽  
Dictionary Learning ◽  
Learning Method ◽  
Task Transfer

scholarly journals Model of Thailand Speech Intelligibility (T-SI) in the Large Classrooms from Public University

2017 ◽  
Vol 13 (7) ◽  
pp. 69
Author(s):  
Pasit Leeniva ◽  
Prapatpong Upala
Keyword(s):  
Speech Intelligibility ◽  
Positive Relationship ◽  
Negative Relationship ◽  
Public University ◽  
Spatial Component ◽  
Surface Areas ◽  
Transmission Index ◽  
Large Classrooms ◽  
Si Model ◽  
Relationship Of

The objectives of this research are to evaluate acoustic environments and to forecast STI values from spatial component variables in the large classrooms of the Thai public university that were specially controlled the same room finishing materials including the floor, walls, and ceiling. Whereas the five spatial component factors included (1) Room Volume (RV), (2) Ceiling Height (CH), (3) the Ratio of Depth to Width (Rdw), (4) Total Room Surface (TS), and (5) Percentage of Absorbing Surface areas (PAS). The research tools were the smartphones that used the applications for acoustical evaluation and speech intelligibility analysis. The Speech Transmission Index (STI), Reverberation Time (RT), and Background Noise Level (BNL) were collected by the calibrated microphone in the nine points distributed across the entire room. And also, the sounds for testing were simulated such as balloon burst, and STIPA signal via a sound generator. The Thailand Speech Intelligibility (T-SI) model was developed by the multiple regression analysis with a statistical at a confidence level of 95%.The results showed that this T-SI model depended on the strongly positive relationship of PAS and the slightly positive relationship of CH, TS while the RV, Rdw were slightly the negative relationship and which predicted STI values. Moreover, the highest affecting variable of T-SI model was CH and the lowest was PAS. However, this research implies that the improving room acoustic quality would be adjusting the sound absorbing surface areas i.e., increase the cloth curtain or appropriate methods.

scholarly journals 2M BeautyNet: Facial Beauty Prediction Based on Multi-Task Transfer Learning

IEEE Access ◽  
2020 ◽  
Vol 8 ◽  
pp. 20245-20256 ◽  
Author(s):  
Junying Gan ◽  
Li Xiang ◽  
Yikui Zhai ◽  
Chaoyun Mai ◽  
Guohui He ◽  
...  
Keyword(s):  
Transfer Learning ◽  
Facial Beauty ◽  
Task Transfer

Milling force prediction model based on transfer learning and neural network

2020 ◽  
Author(s):  
Juncheng Wang ◽  
Bin Zou ◽  
Mingfang Liu ◽  
Yishang Li ◽  
Hongjian Ding ◽  
...  
Keyword(s):  
Neural Network ◽  
Prediction Model ◽  
Transfer Learning ◽  
Milling Force ◽  
Force Prediction ◽  
Model Based

scholarly journals Probabilistic Policy Reuse for inter-task transfer learning

2010 ◽  
Vol 58 (7) ◽  
pp. 866-871 ◽  
Author(s):  
Fernando Fernández ◽  
Javier García ◽  
Manuela Veloso
Keyword(s):  
Transfer Learning ◽  
Task Transfer

scholarly journals Predicting Tyrosine Kinase Inhibitor Treatment Response in Stage IV Lung Adenocarcinoma Patients With EGFR Mutation Using Model-Based Deep Transfer Learning

2021 ◽  
Vol 11 ◽  
Author(s):  
Runping Hou ◽  
Xiaoyang Li ◽  
Junfeng Xiong ◽  
Tianle Shen ◽  
Wen Yu ◽  
...  
Keyword(s):  
Prediction Model ◽  
Transfer Learning ◽  
Clinical Features ◽  
Hybrid Model ◽  
Egfr Mutation ◽  
Stage Iv ◽  
Classification Model ◽  
First Line ◽  
Tki Treatment ◽  
3D Cnn

BackgroundFor stage IV patients harboring EGFR mutations, there is a differential response to the first-line TKI treatment. We constructed three-dimensional convolutional neural networks (CNN) with deep transfer learning to stratify patients into subgroups with different response and progression risks.Materials and MethodsFrom 2013 to 2017, 339 patients with EGFR mutation receiving first-line TKI treatment were included. Progression-free survival (PFS) time and progression patterns were confirmed by routine follow-up and restaging examinations. Patients were divided into two subgroups according to the median PFS (<=9 months, > 9 months). We developed a PFS prediction model and a progression pattern classification model using transfer learning from a pre-trained EGFR mutation classification 3D CNN. Clinical features were fused with the 3D CNN to build the final hybrid prediction model. The performance was quantified using area under receiver operating characteristic curve (AUC), and model performance was compared by AUCs with Delong test.ResultsThe PFS prediction CNN showed an AUC of 0.744 (95% CI, 0.645–0.843) in the independent validation set and the hybrid model of CNNs and clinical features showed an AUC of 0.771 (95% CI, 0.676–0.866), which are significantly better than clinical features-based model (AUC, 0.624, P<0.01). The progression pattern prediction model showed an AUC of 0.762(95% CI, 0.643–0.882) and the hybrid model with clinical features showed an AUC of 0.794 (95% CI, 0.681–0.908), which can provide compensate information for clinical features-based model (AUC, 0.710; 95% CI, 0.582–0.839).ConclusionThe CNN exhibits potential ability to stratify progression status in patients with EGFR mutation treated with first-line TKI, which might help make clinical decisions.

scholarly journals Anti-cancer Drug Synergy Prediction in Understudied Tissues using Transfer Learning

2020 ◽  
Author(s):  
Yejin Kim ◽  
Shuyu Zheng ◽  
Jing Tang ◽  
W. Jim Zheng ◽  
Zhao Li ◽  
...  
Keyword(s):  
Prediction Model ◽  
Transfer Learning ◽  
Treatment Options ◽  
Cancer Drug ◽  
Drug Synergy ◽  
Learning From Data ◽  
Cancer Types ◽  
Improved Accuracy ◽  
Synergy Prediction

AbstractMotivationExploring an exponentially increasing yet more promising space, high-throughput combinatorial drug screening has advantages in identifying cancer treatment options with higher efficacy without degradation in terms of safety. A key challenge is that accumulated number of observations in in-vitro drug responses varies greatly among different cancer types, where some tissues (such as bone and prostate) are understudied than the others. Thus, we aim to develop a drug synergy prediction model for understudied data-poor tissues as overcoming data scarcity problem.ResultsWe collected a comprehensive set of genetic, molecular, phenotypic features for cancer cell lines from six different databases. We developed a drug synergy prediction model based on deep neural networks to integrate multi-modal input and utilize transfer learning from data-rich tissues to data-poor tissues. We showed improved accuracy in predicting drug synergy in understudied tissues without enough drug combination screening data nor after-treatment transcriptome. Our synergy prediction model can be used to rank synergistic drug combinations in understudied tissues and thus help prioritizing future in-vitro experiments.Availability and ImplementationOur algorithm will be publicly available via https://github.com/yejinjkim/drug-synergy-prediction

Sign in / Sign up

Export Citation Format

Share Document