Latent-Space Data Augmentation for Visually-Grounded Language Understanding

2020 ◽  
pp. 179-187
Author(s):  
Aly Magassouba ◽  
Komei Sugiura ◽  
Hisashi Kawai
Keyword(s):  
Data Augmentation ◽  
Language Understanding ◽  
Latent Space ◽  
Space Data

scholarly journals Cooperative Training and Latent Space Data Augmentation for Robust Medical Image Segmentation

2021 ◽  
pp. 149-159
Author(s):  
Chen Chen ◽  
Kerstin Hammernik ◽  
Cheng Ouyang ◽  
Chen Qin ◽  
Wenjia Bai ◽  
...  
Keyword(s):  
Image Segmentation ◽  
Medical Image ◽  
Data Augmentation ◽  
Medical Image Segmentation ◽  
Latent Space ◽  
Cooperative Training ◽  
Space Data

Data Augmentation via Latent Space Interpolation for Image Classification

2018 ◽  
Author(s):  
Xiaofeng Liu ◽  
Yang Zou ◽  
Lingsheng Kong ◽  
Zhihui Diao ◽  
Junliang Yan ◽  
...  
Keyword(s):  
Image Classification ◽  
Data Augmentation ◽  
Latent Space

scholarly journals SpecAugment++: A Hidden Space Data Augmentation Method for AcousticScene Classification

2021 ◽  
Author(s):  
Helin Wang ◽  
Yuexian Zou ◽  
Wenwu Wang
Keyword(s):  
Neural Networks ◽  
Deep Neural Networks ◽  
Data Augmentation ◽  
Input Space ◽  
Space Data ◽  
Time Frames

In this paper, we present SpecAugment++, a novel data aug-mentation method for deep neural networks based acousticscene classification (ASC). Different from other popular dataaugmentation methods such as SpecAugment and mixup thatonly work on the input space, SpecAugment++ is applied toboth the input space and the hidden space of the deep neuralnetworks to enhance the input and the intermediate feature rep-resentations. For an intermediate hidden state, the augmentationtechniques consist of masking blocks of frequency channels andmasking blocks of time frames, which improve generalizationby enabling a model to attend not only to the most discrimina-tive parts of the feature, but also the entire parts. Apart fromusing zeros for masking, we also examine two approaches formasking based on the use of other samples within the mini-batch, which helps introduce noises to the networks to makethem more discriminative for classification. The experimentalresults on the DCASE 2018 Task1 dataset and DCASE 2019Task1 dataset show that our proposed method can obtain3.6%and4.7%accuracy gains over a strong baseline without aug-mentation (i.e.CP-ResNet) respectively, and outperforms otherprevious data augmentation methods.

scholarly journals Automatic CNN-based detection of cardiac MR motion artefacts using k-space data augmentation and curriculum learning

2019 ◽  
Vol 55 ◽  
pp. 136-147 ◽  
Author(s):  
Ilkay Oksuz ◽  
Bram Ruijsink ◽  
Esther Puyol-Antón ◽  
James R. Clough ◽  
Gastao Cruz ◽  
...  
Keyword(s):  
Data Augmentation ◽  
Motion Artefacts ◽  
Cardiac Mr ◽  
Space Data

scholarly journals Structured variational inference for simulating populations of radio galaxies

2021 ◽  
Vol 503 (3) ◽  
pp. 3351-3370
Author(s):  
David J Bastien ◽  
Anna M M Scaife ◽  
Hongming Tang ◽  
Micah Bowles ◽  
Fiona Porter
Keyword(s):  
Data Augmentation ◽  
Radio Galaxies ◽  
Variational Inference ◽  
Postage Stamp ◽  
Synthetic Populations ◽  
Latent Space ◽  
Variational Autoencoder ◽  
Decoder Architecture ◽  
Fully Connected

ABSTRACT We present a model for generating postage stamp images of synthetic Fanaroff–Riley Class I and Class II radio galaxies suitable for use in simulations of future radio surveys such as those being developed for the Square Kilometre Array. This model uses a fully connected neural network to implement structured variational inference through a variational autoencoder and decoder architecture. In order to optimize the dimensionality of the latent space for the autoencoder, we introduce the radio morphology inception score (RAMIS), a quantitative method for assessing the quality of generated images, and discuss in detail how data pre-processing choices can affect the value of this measure. We examine the 2D latent space of the VAEs and discuss how this can be used to control the generation of synthetic populations, whilst also cautioning how it may lead to biases when used for data augmentation.

scholarly journals Latent Space Data Assimilation by using Deep Learning

2021 ◽  
Author(s):  
Mathis Peyron ◽  
Anthony Fillion ◽  
Selime Gürol ◽  
Victor Marchais ◽  
Serge Gratton ◽  
...  
Keyword(s):  
Deep Learning ◽  
Data Assimilation ◽  
Latent Space ◽  
Space Data

scholarly journals Data Augmentation for Spoken Language Understanding via Joint Variational Generation

2019 ◽  
Vol 33 ◽  
pp. 7402-7409 ◽  
Author(s):  
Kang Min Yoo ◽  
Youhyun Shin ◽  
Sang-goo Lee
Keyword(s):  
Latent Variable ◽  
Data Augmentation ◽  
Domain Adaptation ◽  
Latent Variable Models ◽  
Spoken Language ◽  
Generative Models ◽  
Statistical Testing ◽  
Language Understanding ◽  
Data Scarcity ◽  
Spoken Language Understanding

Data scarcity is one of the main obstacles of domain adaptation in spoken language understanding (SLU) due to the high cost of creating manually tagged SLU datasets. Recent works in neural text generative models, particularly latent variable models such as variational autoencoder (VAE), have shown promising results in regards to generating plausible and natural sentences. In this paper, we propose a novel generative architecture which leverages the generative power of latent variable models to jointly synthesize fully annotated utterances. Our experiments show that existing SLU models trained on the additional synthetic examples achieve performance gains. Our approach not only helps alleviate the data scarcity issue in the SLU task for many datasets but also indiscriminately improves language understanding performances for various SLU models, supported by extensive experiments and rigorous statistical testing.

scholarly journals Generative Adversarial Network for Class-Conditional Data Augmentation

2020 ◽  
Vol 10 (23) ◽  
pp. 8415
Author(s):  
Jeongmin Lee ◽  
Younkyoung Yoon ◽  
Junseok Kwon
Keyword(s):  
Classification Accuracy ◽  
Data Augmentation ◽  
Denoising Autoencoder ◽  
Generative Adversarial Network ◽  
Minority Class ◽  
Adversarial Network ◽  
Latent Space ◽  
Benchmark Datasets ◽  
Class Information ◽  
Classification Tasks

We propose a novel generative adversarial network for class-conditional data augmentation (i.e., GANDA) to mitigate data imbalance problems in image classification tasks. The proposed GANDA generates minority class data by exploiting majority class information to enhance the classification accuracy of minority classes. For stable GAN training, we introduce a new denoising autoencoder initialization with explicit class conditioning in the latent space, which enables the generation of definite samples. The generated samples are visually realistic and have a high resolution. Experimental results demonstrate that the proposed GANDA can considerably improve classification accuracy, especially when datasets are highly imbalanced on standard benchmark datasets (i.e., MNIST and CelebA). Our generated samples can be easily used to train conventional classifiers to enhance their classification accuracy.

Sign in / Sign up

Export Citation Format

Share Document