scholarly journals Nonlinear Mixup: Out-Of-Manifold Data Augmentation for Text Classification

2020 ◽  
Vol 34 (04) ◽  
pp. 4044-4051
Author(s):  
Hongyu Guo
Keyword(s):  
Random Sample ◽  
Text Classification ◽  
Data Augmentation ◽  
Convex Combination ◽  
Empirical Studies ◽  
Effective Model ◽  
Training Samples ◽  
Sentence Classification ◽  
Linear Scalar ◽  
Sample Pair

Data augmentation with Mixup (Zhang et al. 2018) has shown to be an effective model regularizer for current art deep classification networks. It generates out-of-manifold samples through linearly interpolating inputs and their corresponding labels of random sample pairs. Despite its great successes, Mixup requires convex combination of the inputs as well as the modeling targets of a sample pair, thus significantly limits the space of its synthetic samples and consequently its regularization effect. To cope with this limitation, we propose “nonlinear Mixup”. Unlike Mixup where the input and label pairs share the same, linear, scalar mixing policy, our approach embraces nonlinear interpolation policy for both the input and label pairs, where the mixing policy for the labels is adaptively learned based on the mixed input. Experiments on benchmark sentence classification datasets indicate that our approach significantly improves upon Mixup. Our empirical studies also show that the out-of-manifold samples generated by our strategy encourage training samples in each class to form a tight representation cluster that is far from others.

scholarly journals Data Augmentation and Spectral Structure Features for Limited Samples Hyperspectral Classification

2021 ◽  
Vol 13 (4) ◽  
pp. 547
Author(s):  
Wenning Wang ◽  
Xuebin Liu ◽  
Xuanqin Mou
Keyword(s):  
Classification Accuracy ◽  
Data Augmentation ◽  
Classification Problem ◽  
Classification Performance ◽  
Spectral Structure ◽  
Limited Sample ◽  
Sample Classification ◽  
Training Samples ◽  
Traditional Classification ◽  
Hyperspectral Classification

For both traditional classification and current popular deep learning methods, the limited sample classification problem is very challenging, and the lack of samples is an important factor affecting the classification performance. Our work includes two aspects. First, the unsupervised data augmentation for all hyperspectral samples not only improves the classification accuracy greatly with the newly added training samples, but also further improves the classification accuracy of the classifier by optimizing the augmented test samples. Second, an effective spectral structure extraction method is designed, and the effective spectral structure features have a better classification accuracy than the true spectral features.

GaitSense: Towards Ubiquitous Gait-Based Human Identification with Wi-Fi

2022 ◽  
Vol 18 (1) ◽  
pp. 1-24
Author(s):  
Yi Zhang ◽  
Yue Zheng ◽  
Guidong Zhang ◽  
Kun Qian ◽  
Chen Qian ◽  
...  
Keyword(s):  
Data Augmentation ◽  
Gait Recognition ◽  
Wearable Sensors ◽  
Human Identification ◽  
Training Data ◽  
Identification Accuracy ◽  
Identification System ◽  
Gait Patterns ◽  
Training Samples ◽  
Augmentation Techniques

Gait, the walking manner of a person, has been perceived as a physical and behavioral trait for human identification. Compared with cameras and wearable sensors, Wi-Fi-based gait recognition is more attractive because Wi-Fi infrastructure is almost available everywhere and is able to sense passively without the requirement of on-body devices. However, existing Wi-Fi sensing approaches impose strong assumptions of fixed user walking trajectories, sufficient training data, and identification of already known users. In this article, we present GaitSense , a Wi-Fi-based human identification system, to overcome the above unrealistic assumptions. To deal with various walking trajectories and speeds, GaitSense first extracts target specific features that best characterize gait patterns and applies novel normalization algorithms to eliminate gait irrelevant perturbation in signals. On this basis, GaitSense reduces the training efforts in new deployment scenarios by transfer learning and data augmentation techniques. GaitSense also enables a distinct feature of illegal user identification by anomaly detection, making the system readily available for real-world deployment. Our implementation and evaluation with commodity Wi-Fi devices demonstrate a consistent identification accuracy across various deployment scenarios with little training samples, pushing the limit of gait recognition with Wi-Fi signals.

scholarly journals Few-Shot Text Classification with Triplet Networks, Data Augmentation, and Curriculum Learning

2021 ◽  
Author(s):  
Jason Wei ◽  
Chengyu Huang ◽  
Soroush Vosoughi ◽  
Yu Cheng ◽  
Shiqi Xu
Keyword(s):  
Text Classification ◽  
Data Augmentation

scholarly journals A Survey on Deep Learning-Based Short/Zero-Calibration Approaches for EEG-Based Brain–Computer Interfaces

2021 ◽  
Vol 15 ◽  
Author(s):  
Wonjun Ko ◽  
Eunjin Jeon ◽  
Seungwoo Jeong ◽  
Jaeun Phyo ◽  
Heung-Il Suk
Keyword(s):  
Deep Learning ◽  
Explicit Knowledge ◽  
Data Augmentation ◽  
Generative Model ◽  
Machine Learning Techniques ◽  
Brain Computer Interfaces ◽  
Computer Interfaces ◽  
Training Samples ◽  
Calibration Methods ◽  
Complex Patterns

Brain–computer interfaces (BCIs) utilizing machine learning techniques are an emerging technology that enables a communication pathway between a user and an external system, such as a computer. Owing to its practicality, electroencephalography (EEG) is one of the most widely used measurements for BCI. However, EEG has complex patterns and EEG-based BCIs mostly involve a cost/time-consuming calibration phase; thus, acquiring sufficient EEG data is rarely possible. Recently, deep learning (DL) has had a theoretical/practical impact on BCI research because of its use in learning representations of complex patterns inherent in EEG. Moreover, algorithmic advances in DL facilitate short/zero-calibration in BCI, thereby suppressing the data acquisition phase. Those advancements include data augmentation (DA), increasing the number of training samples without acquiring additional data, and transfer learning (TL), taking advantage of representative knowledge obtained from one dataset to address the so-called data insufficiency problem in other datasets. In this study, we review DL-based short/zero-calibration methods for BCI. Further, we elaborate methodological/algorithmic trends, highlight intriguing approaches in the literature, and discuss directions for further research. In particular, we search for generative model-based and geometric manipulation-based DA methods. Additionally, we categorize TL techniques in DL-based BCIs into explicit and implicit methods. Our systematization reveals advances in the DA and TL methods. Among the studies reviewed herein, ~45% of DA studies used generative model-based techniques, whereas ~45% of TL studies used explicit knowledge transferring strategy. Moreover, based on our literature review, we recommend an appropriate DA strategy for DL-based BCIs and discuss trends of TLs used in DL-based BCIs.

scholarly journals An End-to-End Learning-Based Row-Following System for an Agricultural Robot in Structured Apple Orchards

2021 ◽  
Vol 2021 ◽  
pp. 1-14
Author(s):  
Peichen Huang ◽  
Lixue Zhu ◽  
Zhigang Zhang ◽  
Chenyu Yang
Keyword(s):  
Navigation System ◽  
Data Augmentation ◽  
Apple Orchard ◽  
Visual Navigation ◽  
Sample Collection ◽  
System A ◽  
Network Training ◽  
Training Samples ◽  
End To End ◽  
Agricultural Robot

A row-following system based on end-to-end learning for an agricultural robot in an apple orchard was developed in this study. Instead of dividing the navigation into multiple traditional subtasks, the designed end-to-end learning method maps images from the camera directly to driving commands, which reduces the complexity of the navigation system. A sample collection method for network training was also proposed, by which the robot could automatically drive and collect data without an operator or remote control. No hand labeling of training samples is required. To improve the network generalization, methods such as batch normalization, dropout, data augmentation, and 10-fold cross-validation were adopted. In addition, internal representations of the network were analyzed, and row-following tests were carried out. Test results showed that the visual navigation system based on end-to-end learning could guide the robot by adjusting its posture according to different scenarios and successfully passing through the tree rows.

scholarly journals Session centered Recommendation Utilizing Future Contexts in Social Media

2021 ◽  
Vol 29 (3) ◽  
pp. 91-104
Author(s):  
Sanjeev Dhawan ◽  
Kulvinder Singh ◽  
Adrian Rabaea ◽  
Amit Batra
Keyword(s):  
Data Augmentation ◽  
Empirical Studies ◽  
Sequential Data ◽  
Space Filling ◽  
The Future ◽  
Future Data ◽  
Real World Datasets ◽  
Sequential Information ◽  
Better Than

Abstract Session centered recommender systems has emerged as an interesting and challenging topic amid researchers during the past few years. In order to make a prediction in the sequential data, prevailing approaches utilize either left to right design autoregressive or data augmentation methods. As these approaches are used to utilize the sequential information pertaining to user conduct, the information about the future context of an objective interaction is totally ignored while making prediction. As a matter of fact, we claim that during the course of training, the future data after the objective interaction are present and this supplies indispensable signal on preferences of users and if utilized can increase the quality of recommendation. It is a subtle task to incorporate future contexts into the process of training, as the rules of machine learning are not followed and can result in loss of data. Therefore, in order to solve this problem, we suggest a novel encoder decoder prototype termed as space filling centered Recommender (SRec), which is used to train the encoder and decoder utilizing space filling approach. Particularly, an incomplete sequence is taken into consideration by the encoder as input (few items are absent) and then decoder is used to predict these items which are absent initially based on the encoded interpretation. The general SRec prototype is instantiated by us employing convolutional neural network (CNN) by giving emphasis on both e ciency and accuracy. The empirical studies and investigation on two real world datasets are conducted by us including short, medium and long sequences, which exhibits that SRec performs better than traditional sequential recommendation approaches.

Sign in / Sign up

Export Citation Format

Share Document