Browsing by Subject "Training data"

Now showing 1 - 4 of 4
  • Thumbnail Image
    Item
    Energy-Efficient Device Selection in Federated Edge Learning
    (IEEE, 2021-07) Peng, Cheng; Hu, Qin; Chen, Jianan; Kang, Kyubyung; Li, Feng; Zou, Xukai; Computer and Information Science, School of Science
    Due to the increasing demand from mobile devices for the real-time response of cloud computing services, federated edge learning (FEL) emerges as a new computing paradigm, which utilizes edge devices to achieve efficient machine learning while protecting their data privacy. Implementing efficient FEL suffers from the challenges of devices’ limited computing and communication resources, as well as unevenly distributed datasets, which inspires several existing research focusing on device selection to optimize time consumption and data diversity. However, these studies fail to consider the energy consumption of edge devices given their limited power supply, which can seriously affect the cost-efficiency of FEL with unexpected device dropouts. To fill this gap, we propose a device selection model capturing both energy consumption and data diversity optimization, under the constraints of time consumption and training data amount. Then we solve the optimization problem by reformulating the original model and designing a novel algorithm, named E2DS, to reduce the time complexity greatly. By comparing with two classical FEL schemes, we validate the superiority of our proposed device selection mechanism for FEL with extensive experimental results.
  • Thumbnail Image
    Item
    Improved Adverse Drug Event Prediction Through Information Component Guided Pharmacological Network Model (IC-PNM)
    (IEEE, 2021) Ji, Xiangmin; Wang, Lei; Hua, Liyan; Wang, Xueying; Zhang, Pengyue; Shendre, Aditi; Feng, Weixing; Li, Jin; Li, Lang; Biostatistics and Health Data Science, Richard M. Fairbanks School of Public Health
    Improving adverse drug event (ADE) prediction is highly critical in pharmacovigilance research. We propose a novel information component guided pharmacological network model (IC-PNM) to predict drug-ADE signals. This new method combines the pharmacological network model and information component, a Bayes statistics method. We use 33,947 drug-ADE pairs from the FDA Adverse Event Reporting System (FAERS) 2010 data as the training data, and the new 21,065 drug-ADE pairs from FAERS 2011-2015 as the validations samples. The IC-PNM data analysis suggests that both large and small sample size drug-ADE pairs are needed in training the predictive model for its prediction performance to reach an area under the receiver operating characteristic curve (\textAUROC)= 0.82(AUROC)=0.82. On the other hand, the IC-PNM prediction performance improved to \textAUROC= 0.91AUROC=0.91 if we removed the small sample size drug-ADE pairs from the prediction model during validation.
  • Thumbnail Image
    Item
    mmFit: Low-Effort Personalized Fitness Monitoring Using Millimeter Wave
    (IEEE, 2022) Xie, Yucheng; Jiang, Ruizhe; Guo, Xiaonan; Wang, Yan; Cheng, Jerry; Chen, Yingying; Electrical and Computer Engineering, Purdue School of Engineering and Technology
    There is a growing trend for people to perform work-outs at home due to the global pandemic of COVID-19 and the stay-at-home policy of many countries. Since a self-designed fitness plan often lacks professional guidance to achieve ideal outcomes, it is important to have an in-home fitness monitoring system that can track the exercise process of users. Traditional camera-based fitness monitoring may raise serious privacy concerns, while sensor-based methods require users to wear dedicated devices. Recently, researchers propose to utilize RF signals to enable non-intrusive fitness monitoring, but these approaches all require huge training efforts from users to achieve a satisfactory performance, especially when the system is used by multiple users (e.g., family members). In this work, we design and implement a fitness monitoring system using a single COTS mm Wave device. The proposed system integrates workout recognition, user identification, multi-user monitoring, and training effort reduction modules and makes them work together in a single system. In particular, we develop a domain adaptation framework to reduce the amount of training data collected from different domains via mitigating impacts caused by domain characteristics embedded in mm Wave signals. We also develop a GAN-assisted method to achieve better user identification and workout recognition when only limited training data from the same domain is available. We propose a unique spatialtemporal heatmap feature to achieve personalized workout recognition and develop a clustering-based method for concurrent workout monitoring. Extensive experiments with 14 typical workouts involving 11 participants demonstrate that our system can achieve 97% average workout recognition accuracy and 91% user identification accuracy.
  • Thumbnail Image
    Item
    Towards Fair Cross-Domain Adaptation via Generative Learning
    (IEEE, 2021) Wang, Tongxin; Ding, Zhengming; Shao, Wei; Tang, Haixu; Huang, Kun; Medicine, School of Medicine
    Domain Adaptation (DA) targets at adapting a model trained over the well-labeled source domain to the unlabeled target domain lying in different distributions. Existing DA normally assumes the well-labeled source domain is class-wise balanced, which means the size per source class is relatively similar. However, in real-world applications, labeled samples for some categories in the source domain could be extremely few due to the difficulty of data collection and annotation, which leads to decreasing performance over target domain on those few-shot categories. To perform fair cross-domain adaptation and boost the performance on these minority categories, we develop a novel Generative Few-shot Cross-domain Adaptation (GFCA) algorithm for fair cross-domain classification. Specifically, generative feature augmentation is explored to synthesize effective training data for few-shot source classes, while effective cross-domain alignment aims to adapt knowledge from source to facilitate the target learning. Experimental results on two large cross-domain visual datasets demonstrate the effectiveness of our proposed method on improving both few-shot and overall classification accuracy comparing with the state-of-the-art DA approaches.