scholarly journals Improving the resolution of migrated images by approximating the inverse Hessian using deep learning

Geophysics ◽  
2020 ◽  
Vol 85 (4) ◽  
pp. WA173-WA183 ◽  
Author(s):  
Harpreet Kaur ◽  
Nam Pham ◽  
Sergey Fomel
Keyword(s):  
Field Data ◽  
Velocity Model ◽  
Image Resolution ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
Data Sets ◽  
Computationally Efficient ◽  
Validation Data ◽  
Network Output

We have estimated migrated images with meaningful amplitudes matching least-squares migrated images by approximating the inverse Hessian using generative adversarial networks (GANs) in a conditional setting. We use the CycleGAN framework and extend it to the conditional CycleGAN such that the mapping from the migrated image to the true reflectivity is subjected to a velocity attribute condition. This algorithm is applied after migration and is computationally efficient. It produces results comparable to iterative inversion but at a significantly reduced cost. In numerical experiments with synthetic and field data sets, the adopted method improves image resolution, attenuates noise, reduces migration artifacts, and enhances reflection amplitudes. We train the network with three different data sets and test on three other data sets, which are not a part of training. Tests on validation data sets verify the effectiveness of the approach. In addition, the field-data example also highlights the effect of the bandwidth of the training data and the quality of the velocity model on the quality of the deep neural network output.

SaltSeg: Automatic 3D salt segmentation using a deep convolutional neural network

2019 ◽  
Vol 7 (3) ◽  
pp. SE113-SE122 ◽  
Author(s):  
Yunzhi Shi ◽  
Xinming Wu ◽  
Sergey Fomel
Keyword(s):  
Large Scale ◽  
Model Building ◽  
Ground Truth ◽  
Velocity Model ◽  
Training Data ◽  
Data Sets ◽  
Validation Data ◽  
Data Set ◽  
Seismic Image ◽  
Data Generator

Salt boundary interpretation is important for the understanding of salt tectonics and velocity model building for seismic migration. Conventional methods consist of computing salt attributes and extracting salt boundaries. We have formulated the problem as 3D image segmentation and evaluated an efficient approach based on deep convolutional neural networks (CNNs) with an encoder-decoder architecture. To train the model, we design a data generator that extracts randomly positioned subvolumes from large-scale 3D training data set followed by data augmentation, then feed a large number of subvolumes into the network while using salt/nonsalt binary labels generated by thresholding the velocity model as ground truth labels. We test the model on validation data sets and compare the blind test predictions with the ground truth. Our results indicate that our method is capable of automatically capturing subtle salt features from the 3D seismic image with less or no need for manual input. We further test the model on a field example to indicate the generalization of this deep CNN method across different data sets.

scholarly journals Monte Carlo and Reconstruction Membership Inference Attacks against Generative Models

2019 ◽  
Vol 2019 (4) ◽  
pp. 232-249 ◽  
Author(s):  
Benjamin Hilprecht ◽  
Martin Härterich ◽  
Daniel Bernau
Keyword(s):  
Data Privacy ◽  
Information Leakage ◽  
Generative Models ◽  
Generative Model ◽  
Training Data ◽  
Generative Adversarial Networks ◽  
Data Sets ◽  
Success Rates ◽  
Model Quality ◽  
Type Formalization

Abstract We present two information leakage attacks that outperform previous work on membership inference against generative models. The first attack allows membership inference without assumptions on the type of the generative model. Contrary to previous evaluation metrics for generative models, like Kernel Density Estimation, it only considers samples of the model which are close to training data records. The second attack specifically targets Variational Autoencoders, achieving high membership inference accuracy. Furthermore, previous work mostly considers membership inference adversaries who perform single record membership inference. We argue for considering regulatory actors who perform set membership inference to identify the use of specific datasets for training. The attacks are evaluated on two generative model architectures, Generative Adversarial Networks (GANs) and Variational Autoen-coders (VAEs), trained on standard image datasets. Our results show that the two attacks yield success rates superior to previous work on most data sets while at the same time having only very mild assumptions. We envision the two attacks in combination with the membership inference attack type formalization as especially useful. For example, to enforce data privacy standards and automatically assessing model quality in machine learning as a service setups. In practice, our work motivates the use of GANs since they prove less vulnerable against information leakage attacks while producing detailed samples.

De-noising of transient electromagnetic data based on the long short-term memory-autoencoder

2020 ◽  
Vol 224 (1) ◽  
pp. 669-681
Author(s):  
Sihong Wu ◽  
Qinghua Huang ◽  
Li Zhao
Keyword(s):  
Field Data ◽  
Short Term Memory ◽  
Noise Removal ◽  
Data Sets ◽  
Late Time ◽  
Short Term ◽  
Term Memory ◽  
Transient Electromagnetic ◽  
Long Short Term Memory

SUMMARY Late-time transient electromagnetic (TEM) data contain deep subsurface information and are important for resolving deeper electrical structures. However, due to their relatively small signal amplitudes, TEM responses later in time are often dominated by ambient noises. Therefore, noise removal is critical to the application of TEM data in imaging electrical structures at depth. De-noising techniques for TEM data have been developed rapidly in recent years. Although strong efforts have been made to improving the quality of the TEM responses, it is still a challenge to effectively extract the signals due to unpredictable and irregular noises. In this study, we develop a new type of neural network architecture by combining the long short-term memory (LSTM) network with the autoencoder structure to suppress noise in TEM signals. The resulting LSTM-autoencoders yield excellent performance on synthetic data sets including horizontal components of the electric field and vertical component of the magnetic field generated by different sources such as dipole, loop and grounded line sources. The relative errors between the de-noised data sets and the corresponding noise-free transients are below 1% for most of the sampling points. Notable improvement in the resistivity structure inversion result is achieved using the TEM data de-noised by the LSTM-autoencoder in comparison with several widely-used neural networks, especially for later-arriving signals that are important for constraining deeper structures. We demonstrate the effectiveness and general applicability of the LSTM-autoencoder by de-noising experiments using synthetic 1-D and 3-D TEM signals as well as field data sets. The field data from a fixed loop survey using multiple receivers are greatly improved after de-noising by the LSTM-autoencoder, resulting in more consistent inversion models with significantly increased exploration depth. The LSTM-autoencoder is capable of enhancing the quality of the TEM signals at later times, which enables us to better resolve deeper electrical structures.

LDpred-funct: incorporating functional priors improves polygenic prediction accuracy in UK Biobank and 23andMe data sets

2018 ◽  
Author(s):  
Carla Márquez-Luna ◽  
Steven Gazal ◽  
Po-Ru Loh ◽  
Samuel S. Kim ◽  
Nicholas Furlotte ◽  
...  
Keyword(s):  
Complex Traits ◽  
Prediction Accuracy ◽  
Causal Effect ◽  
Complex Trait ◽  
Training Data ◽  
Data Sets ◽  
Uk Biobank ◽  
Validation Data ◽  
Functional Regions ◽  
The Uk

AbstractGenetic variants in functional regions of the genome are enriched for complex trait heritability. Here, we introduce a new method for polygenic prediction, LDpred-funct, that leverages trait-specific functional priors to increase prediction accuracy. We fit priors using the recently developed baseline-LD model, which includes coding, conserved, regulatory and LD-related annotations. We analytically estimate posterior mean causal effect sizes and then use cross-validation to regularize these estimates, improving prediction accuracy for sparse architectures. LDpred-funct attained higher prediction accuracy than other polygenic prediction methods in simulations using real genotypes. We applied LDpred-funct to predict 21 highly heritable traits in the UK Biobank. We used association statistics from British-ancestry samples as training data (avg N=373K) and samples of other European ancestries as validation data (avg N=22K), to minimize confounding. LDpred-funct attained a +4.6% relative improvement in average prediction accuracy (avg prediction R2=0.144; highest R2=0.413 for height) compared to SBayesR (the best method that does not incorporate functional information). For height, meta-analyzing training data from UK Biobank and 23andMe cohorts (total N=1107K; higher heritability in UK Biobank cohort) increased prediction R2 to 0.431. Our results show that incorporating functional priors improves polygenic prediction accuracy, consistent with the functional architecture of complex traits.

Saliency Optimization and Integration Via Iterative Bootstrap Learning

2018 ◽  
Vol 32 (09) ◽  
pp. 1859016
Author(s):  
Liming Li ◽  
Xiaodong Chai ◽  
Shuguang Zhao ◽  
Shubin Zheng ◽  
Shengchao Su
Keyword(s):  
Multiple Scales ◽  
Saliency Detection ◽  
Saliency Map ◽  
Input Image ◽  
Training Data ◽  
Data Sets ◽  
Saliency Maps ◽  
Training Samples ◽  
Public Data

This paper proposes an effective method to elevate the performance of saliency detection via iterative bootstrap learning, which consists of two tasks including saliency optimization and saliency integration. Specifically, first, multiscale segmentation and feature extraction are performed on the input image successively. Second, prior saliency maps are generated using existing saliency models, which are used to generate the initial saliency map. Third, prior maps are fed into the saliency regressor together, where training samples are collected from the prior maps at multiple scales and the random forest regressor is learned from such training data. An integration of the initial saliency map and the output of saliency regressor is deployed to generate the coarse saliency map. Finally, in order to improve the quality of saliency map further, both initial and coarse saliency maps are fed into the saliency regressor together, and then the output of the saliency regressor, the initial saliency map as well as the coarse saliency map are integrated into the final saliency map. Experimental results on three public data sets demonstrate that the proposed method consistently achieves the best performance and significant improvement can be obtained when applying our method to existing saliency models.

scholarly journals Completing the ENCODE3 compendium yields accurate imputations across a variety of assays and human biosamples

2019 ◽  
Author(s):  
Jacob Schreiber ◽  
Jeffrey Bilmes ◽  
William Stafford Noble
Keyword(s):  
Biological Activity ◽  
Protein Binding ◽  
Histone Modification ◽  
Chromatin Accessibility ◽  
Training Data ◽  
Data Sets ◽  
Cellular Mechanisms ◽  
Data Set ◽  
Genome Wide

AbstractMotivationRecent efforts to describe the human epigenome have yielded thousands of uniformly processed epigenomic and transcriptomic data sets. These data sets characterize a rich variety of biological activity in hundreds of human cell lines and tissues (“biosamples”). Understanding these data sets, and specifically how they differ across biosamples, can help explain many cellular mechanisms, particularly those driving development and disease. However, due primarily to cost, the total number of assays that can be performed is limited. Previously described imputation approaches, such as Avocado, have sought to overcome this limitation by predicting genome-wide epigenomics experiments using learned associations among available epigenomic data sets. However, these previous imputations have focused primarily on measurements of histone modification and chromatin accessibility, despite other biological activity being crucially important.ResultsWe applied Avocado to a data set of 3,814 tracks of data derived from the ENCODE compendium, spanning 400 human biosamples and 84 assays. The resulting imputations cover measurements of chromatin accessibility, histone modification, transcription, and protein binding. We demonstrate the quality of these imputations by comprehensively evaluating the model’s predictions and by showing significant improvements in protein binding performance compared to the top models in an ENCODE-DREAM challenge. Additionally, we show that the Avocado model allows for efficient addition of new assays and biosamples to a pre-trained model, achieving high accuracy at predicting protein binding, even with only a single track of training data.AvailabilityTutorials and source code are available under an Apache 2.0 license at https://github.com/jmschrei/[email protected] or [email protected]

scholarly journals 3D fluorescence microscopy data synthesis for segmentation and benchmarking

PLoS ONE ◽  
2021 ◽  
Vol 16 (12) ◽  
pp. e0260509
Author(s):  
Dennis Eschweiler ◽  
Malte Rethwisch ◽  
Mareike Jarchow ◽  
Simon Koppers ◽  
Johannes Stegmaier
Keyword(s):  
Fluorescence Microscopy ◽  
Image Data ◽  
Training Data ◽  
Cellular Structures ◽  
Generative Adversarial Networks ◽  
Data Sets ◽  
Automated Generation ◽  
3D Fluorescence Microscopy ◽  
Manual Interaction ◽  
Microscopy Data

Automated image processing approaches are indispensable for many biomedical experiments and help to cope with the increasing amount of microscopy image data in a fast and reproducible way. Especially state-of-the-art deep learning-based approaches most often require large amounts of annotated training data to produce accurate and generalist outputs, but they are often compromised by the general lack of those annotated data sets. In this work, we propose how conditional generative adversarial networks can be utilized to generate realistic image data for 3D fluorescence microscopy from annotation masks of 3D cellular structures. In combination with mask simulation approaches, we demonstrate the generation of fully-annotated 3D microscopy data sets that we make publicly available for training or benchmarking. An additional positional conditioning of the cellular structures enables the reconstruction of position-dependent intensity characteristics and allows to generate image data of different quality levels. A patch-wise working principle and a subsequent full-size reassemble strategy is used to generate image data of arbitrary size and different organisms. We present this as a proof-of-concept for the automated generation of fully-annotated training data sets requiring only a minimum of manual interaction to alleviate the need of manual annotations.

scholarly journals A review and framework for the evaluation of pixel-level uncertainty estimates in satellite aerosol remote sensing

2020 ◽  
Vol 13 (2) ◽  
pp. 373-404 ◽  
Author(s):  
Andrew M. Sayer ◽  
Yves Govaerts ◽  
Pekka Kolmonen ◽  
Antti Lipponen ◽  
Marta Luffarelli ◽  
...  
Keyword(s):  
Uncertainty Propagation ◽  
Optimal Estimation ◽  
Sensitivity Analyses ◽  
Data Sets ◽  
Validation Data ◽  
Data Set ◽  
Formal Error ◽  
Uncertainty Estimates ◽  
Diagnostic Approaches

Abstract. Recent years have seen the increasing inclusion of per-retrieval prognostic (predictive) uncertainty estimates within satellite aerosol optical depth (AOD) data sets, providing users with quantitative tools to assist in the optimal use of these data. Prognostic estimates contrast with diagnostic (i.e. relative to some external truth) ones, which are typically obtained using sensitivity and/or validation analyses. Up to now, however, the quality of these uncertainty estimates has not been routinely assessed. This study presents a review of existing prognostic and diagnostic approaches for quantifying uncertainty in satellite AOD retrievals, and it presents a general framework to evaluate them based on the expected statistical properties of ensembles of estimated uncertainties and actual retrieval errors. It is hoped that this framework will be adopted as a complement to existing AOD validation exercises; it is not restricted to AOD and can in principle be applied to other quantities for which a reference validation data set is available. This framework is then applied to assess the uncertainties provided by several satellite data sets (seven over land, five over water), which draw on methods from the empirical to sensitivity analyses to formal error propagation, at 12 Aerosol Robotic Network (AERONET) sites. The AERONET sites are divided into those for which it is expected that the techniques will perform well and those for which some complexity about the site may provide a more severe test. Overall, all techniques show some skill in that larger estimated uncertainties are generally associated with larger observed errors, although they are sometimes poorly calibrated (i.e. too small or too large in magnitude). No technique uniformly performs best. For powerful formal uncertainty propagation approaches such as optimal estimation, the results illustrate some of the difficulties in appropriate population of the covariance matrices required by the technique. When the data sets are confronted by a situation strongly counter to the retrieval forward model (e.g. potentially mixed land–water surfaces or aerosol optical properties outside the family of assumptions), some algorithms fail to provide a retrieval, while others do but with a quantitatively unreliable uncertainty estimate. The discussion suggests paths forward for the refinement of these techniques.

scholarly journals Prediction of Wine Sensorial Quality by Routinely Measured Chemical Properties

2014 ◽  
Vol 13 (2) ◽  
pp. 182-196
Author(s):  
Adriána Bednárová ◽  
Roman Kranvogl ◽  
Darinka Brodnjak-Vončina ◽  
Tjaša Jug
Keyword(s):  
Chemical Properties ◽  
Principal Component ◽  
Training Data ◽  
Coefficient Of Determination ◽  
Validation Data ◽  
Chemical Descriptors ◽  
Sensorial Quality ◽  
One Year

Abstract The determination of the sensorial quality of wines is of great interest for wine consumers and producers since it declares the quality in most of the cases. The sensorial assays carried out by a group of experts are time-consuming and expensive especially when dealing with large batches of wines. Therefore, an attempt was made to assess the possibility of estimating the wine sensorial quality with using routinely measured chemical descriptors as predictors. For this purpose, 131 Slovenian red wine samples of different varieties and years of production were analysed and correlation and principal component analysis were applied to find inter-relations between the studied oenological descriptors. The method of artificial neural networks (ANNs) was utilised as the prediction tool for estimating overall sensorial quality of red wines. Each model was rigorously validated and sensitivity analysis was applied as a method for selecting the most important predictors. Consequently, acceptable results were obtained, when data representing only one year of production were included in the analysis. In this case, the coefficient of determination (R2) associated with training data was 0.95 and that for validation data was 0.90. When estimating sensorial quality in categorical form, 94 % and 85 % of correctly classified samples were achieved for training and validation subset, respectively.

Sign in / Sign up

Export Citation Format

Share Document