scholarly journals Uncertainty-Aware Deep Classifiers Using Generative Models

2020 ◽  
Vol 34 (04) ◽  
pp. 5620-5627 ◽  
Author(s):  
Murat Sensoy ◽  
Lance Kaplan ◽  
Federico Cerutti ◽  
Maryam Saleki
Keyword(s):  
Neural Networks ◽  
Epistemic Uncertainty ◽  
Feature Space ◽  
Generative Models ◽  
Detection Methods ◽  
Generative Adversarial Networks ◽  
Data Sets ◽  
Bayesian Approaches ◽  
Data Set ◽  
Auxiliary Data

Deep neural networks are often ignorant about what they do not know and overconfident when they make uninformed predictions. Some recent approaches quantify classification uncertainty directly by training the model to output high uncertainty for the data samples close to class boundaries or from the outside of the training distribution. These approaches use an auxiliary data set during training to represent out-of-distribution samples. However, selection or creation of such an auxiliary data set is non-trivial, especially for high dimensional data such as images. In this work we develop a novel neural network model that is able to express both aleatoric and epistemic uncertainty to distinguish decision boundary and out-of-distribution regions of the feature space. To this end, variational autoencoders and generative adversarial networks are incorporated to automatically generate out-of-distribution exemplars for training. Through extensive analysis, we demonstrate that the proposed approach provides better estimates of uncertainty for in- and out-of-distribution samples, and adversarial examples on well-known data sets against state-of-the-art approaches including recent Bayesian approaches for neural networks and anomaly detection methods.

scholarly journals Evaluation of Image Forgery Detection Using Multi-Scale Weber Local Descriptors

2015 ◽  
Vol 24 (04) ◽  
pp. 1540016 ◽  
Author(s):  
Muhammad Hussain ◽  
Sahar Qasem ◽  
George Bebis ◽  
Ghulam Muhammad ◽  
Hatim Aboalsamh ◽  
...  
Keyword(s):  
Digital Image ◽  
Image Data ◽  
Feature Space ◽  
Detection Methods ◽  
Support Vector ◽  
Data Sets ◽  
Forgery Detection ◽  
Data Set ◽  
Multi Scale ◽  
Copy Move Forgery Detection

Due to the maturing of digital image processing techniques, there are many tools that can forge an image easily without leaving visible traces and lead to the problem of the authentication of digital images. Based on the assumption that forgery alters the texture micro-patterns in a digital image and texture descriptors can be used for modeling this change; we employed two stat-of-the-art local texture descriptors: multi-scale Weber's law descriptor (multi-WLD) and multi-scale local binary pattern (multi-LBP) for splicing and copy-move forgery detection. As the tamper traces are not visible to open eyes, so the chrominance components of an image encode these traces and were used for modeling tamper traces with the texture descriptors. To reduce the dimension of the feature space and get rid of redundant features, we employed locally learning based (LLB) algorithm. For identifying an image as authentic or tampered, Support vector machine (SVM) was used. This paper presents the thorough investigation for the validation of this forgery detection method. The experiments were conducted on three benchmark image data sets, namely, CASIA v1.0, CASIA v2.0, and Columbia color. The experimental results showed that the accuracy rate of multi-WLD based method was 94.19% on CASIA v1.0, 96.52% on CASIA v2.0, and 94.17% on Columbia data set. It is not only significantly better than multi-LBP based method, but also it outperforms other stat-of-the-art similar forgery detection methods.

scholarly journals Image Colorization Progress: A Review of Deep Learning Techniques for Automation of Colorization

2021 ◽  
Vol 10 (4) ◽  
pp. 2908-2915
Keyword(s):  
Neural Networks ◽  
Generative Models ◽  
Generative Adversarial Networks ◽  
Data Sets ◽  
Image Colorization ◽  
Black And White ◽  
The Past ◽  
Adversarial Networks ◽  
Learning Techniques ◽  
Made In

Image colorization is the process of taking an input gray- scale (black and white) image and then producing an output colorized image that represents the semantic color tones of the input. Since the past few years, the process of automatic image colorization has been of significant interest and a lot of progress has been made in the field by various researchers. Image colorization finds its application in many domains including medical imaging, restoration of historical documents, etc. There have been different approaches to solve this problem using Convolutional Neural Networks as well as Generative Adversarial Networks. These colorization networks are not only based on different architectures but also are tested on varied data sets. This paper aims to cover some of these proposed approaches through different techniques. The results between the generative models and traditional deep neural networks are compared along with presenting the current limitations in those. The paper proposes a summarized view of past and current advances in the field of image colorization contributed by different authors and researchers.

scholarly journals Comparison of Novelty Detection Methods for Detection of Various Rotary Machinery Faults

Sensors ◽  
2021 ◽  
Vol 21 (10) ◽  
pp. 3536
Author(s):  
Jakub Górski ◽  
Adam Jabłoński ◽  
Mateusz Heesch ◽  
Michał Dziendzikowski ◽  
Ziemowit Dworakowski
Keyword(s):  
Novelty Detection ◽  
Feature Space ◽  
Detection Methods ◽  
Data Sets ◽  
Similar Distribution ◽  
Detection Algorithms ◽  
Rotary Machinery ◽  
Detection Approach ◽  
Input Reconstruction ◽  
Quantitative Indicators

Condition monitoring is an indispensable element related to the operation of rotating machinery. In this article, the monitoring system for the parallel gearbox was proposed. The novelty detection approach is used to develop the condition assessment support system, which requires data collection for a healthy structure. The measured signals were processed to extract quantitative indicators sensitive to the type of damage occurring in this type of structure. The indicator’s values were used for the development of four different novelty detection algorithms. Presented novelty detection models operate on three principles: feature space distance, probability distribution, and input reconstruction. One of the distance-based models is adaptive, adjusting to new data flowing in the form of a stream. The authors test the developed algorithms on experimental and simulation data with a similar distribution, using the training set consisting mainly of samples generated by the simulator. Presented in the article results demonstrate the effectiveness of the trained models on both data sets.

scholarly journals Generation of geometric interpolations of building types with deep variational autoencoders

2020 ◽  
Vol 6 ◽  
Author(s):  
Jaime de Miguel Rodríguez ◽  
Maria Eugenia Villafañe ◽  
Luka Piškorec ◽  
Fernando Sancho Caparrini
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Large Data ◽  
Learning Model ◽  
Large Data Sets ◽  
Data Sets ◽  
Connectivity Map ◽  
Data Set ◽  
3D Objects ◽  
Machine Learning Model

Abstract This work presents a methodology for the generation of novel 3D objects resembling wireframes of building types. These result from the reconstruction of interpolated locations within the learnt distribution of variational autoencoders (VAEs), a deep generative machine learning model based on neural networks. The data set used features a scheme for geometry representation based on a ‘connectivity map’ that is especially suited to express the wireframe objects that compose it. Additionally, the input samples are generated through ‘parametric augmentation’, a strategy proposed in this study that creates coherent variations among data by enabling a set of parameters to alter representative features on a given building type. In the experiments that are described in this paper, more than 150 k input samples belonging to two building types have been processed during the training of a VAE model. The main contribution of this paper has been to explore parametric augmentation for the generation of large data sets of 3D geometries, showcasing its problems and limitations in the context of neural networks and VAEs. Results show that the generation of interpolated hybrid geometries is a challenging task. Despite the difficulty of the endeavour, promising advances are presented.

scholarly journals Exploiting node metadata to predict interactions in large networks using graph embedding and neural networks

2021 ◽  
Author(s):  
Rogini Runghen ◽  
Daniel B Stouffer ◽  
Giulio Valentino Dalla Riva
Keyword(s):  
Machine Learning ◽  
Neural Networks ◽  
Link Prediction ◽  
Graph Embedding ◽  
Feature Space ◽  
Machine Learning Techniques ◽  
Large Networks ◽  
Data Set ◽  
Learning Techniques ◽  
Low Dimensional

Collecting network interaction data is difficult. Non-exhaustive sampling and complex hidden processes often result in an incomplete data set. Thus, identifying potentially present but unobserved interactions is crucial both in understanding the structure of large scale data, and in predicting how previously unseen elements will interact. Recent studies in network analysis have shown that accounting for metadata (such as node attributes) can improve both our understanding of how nodes interact with one another, and the accuracy of link prediction. However, the dimension of the object we need to learn to predict interactions in a network grows quickly with the number of nodes. Therefore, it becomes computationally and conceptually challenging for large networks. Here, we present a new predictive procedure combining a graph embedding method with machine learning techniques to predict interactions on the base of nodes' metadata. Graph embedding methods project the nodes of a network onto a---low dimensional---latent feature space. The position of the nodes in the latent feature space can then be used to predict interactions between nodes. Learning a mapping of the nodes' metadata to their position in a latent feature space corresponds to a classic---and low dimensional---machine learning problem. In our current study we used the Random Dot Product Graph model to estimate the embedding of an observed network, and we tested different neural networks architectures to predict the position of nodes in the latent feature space. Flexible machine learning techniques to map the nodes onto their latent positions allow to account for multivariate and possibly complex nodes' metadata. To illustrate the utility of the proposed procedure, we apply it to a large dataset of tourist visits to destinations across New Zealand. We found that our procedure accurately predicts interactions for both existing nodes and nodes newly added to the network, while being computationally feasible even for very large networks. Overall, our study highlights that by exploiting the properties of a well understood statistical model for complex networks and combining it with standard machine learning techniques, we can simplify the link prediction problem when incorporating multivariate node metadata. Our procedure can be immediately applied to different types of networks, and to a wide variety of data from different systems. As such, both from a network science and data science perspective, our work offers a flexible and generalisable procedure for link prediction.

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.

Simple Convolutional-Based Models: Are They Learning the Task or the Data?

2021 ◽  
pp. 1-17
Author(s):  
Luis Sa-Couto ◽  
Andreas Wichert
Keyword(s):  
Neural Networks ◽  
Pattern Recognition ◽  
Training Data ◽  
Model Complexity ◽  
Data Sets ◽  
Simple Task ◽  
Data Set ◽  
Knowing That ◽  
Handwritten Digit ◽  
End To End

Abstract Convolutional neural networks (CNNs) evolved from Fukushima's neocognitron model, which is based on the ideas of Hubel and Wiesel about the early stages of the visual cortex. Unlike other branches of neocognitron-based models, the typical CNN is based on end-to-end supervised learning by backpropagation and removes the focus from built-in invariance mechanisms, using pooling not as a way to tolerate small shifts but as a regularization tool that decreases model complexity. These properties of end-to-end supervision and flexibility of structure allow the typical CNN to become highly tuned to the training data, leading to extremely high accuracies on typical visual pattern recognition data sets. However, in this work, we hypothesize that there is a flip side to this capability, a hidden overfitting. More concretely, a supervised, backpropagation based CNN will outperform a neocognitron/map transformation cascade (MTCCXC) when trained and tested inside the same data set. Yet if we take both models trained and test them on the same task but on another data set (without retraining), the overfitting appears. Other neocognitron descendants like the What-Where model go in a different direction. In these models, learning remains unsupervised, but more structure is added to capture invariance to typical changes. Knowing that, we further hypothesize that if we repeat the same experiments with this model, the lack of supervision may make it worse than the typical CNN inside the same data set, but the added structure will make it generalize even better to another one. To put our hypothesis to the test, we choose the simple task of handwritten digit classification and take two well-known data sets of it: MNIST and ETL-1. To try to make the two data sets as similar as possible, we experiment with several types of preprocessing. However, regardless of the type in question, the results align exactly with expectation.

A scalable eigenspace-based fuzzy c-means for topic detection

2021 ◽  
Vol ahead-of-print (ahead-of-print) ◽  
Author(s):  
Hendri Murfi
Keyword(s):  
Representation Learning ◽  
Detection Methods ◽  
Data Sets ◽  
Topic Detection ◽  
Data Set ◽  
Content Type ◽  
Running Time ◽  
Fuzzy C Means ◽  
Coherence Score ◽  
Value Decomposition

PurposeThe aim of this research is to develop an eigenspace-based fuzzy c-means method for scalable topic detection.Design/methodology/approachThe eigenspace-based fuzzy c-means (EFCM) combines representation learning and clustering. The textual data are transformed into a lower-dimensional eigenspace using truncated singular value decomposition. Fuzzy c-means is performed on the eigenspace to identify the centroids of each cluster. The topics are provided by transforming back the centroids into the nonnegative subspace of the original space. In this paper, we extend the EFCM method for scalability by using the two approaches, i.e. single-pass and online. We call the developed topic detection methods as oEFCM and spEFCM.FindingsOur simulation shows that both oEFCM and spEFCM methods provide faster running times than EFCM for data sets that do not fit in memory. However, there is a decrease in the average coherence score. For both data sets that fit and do not fit into memory, the oEFCM method provides a tradeoff between running time and coherence score, which is better than spEFCM.Originality/valueThis research produces a scalable topic detection method. Besides this scalability capability, the developed method also provides a faster running time for the data set that fits in memory.

A systematical approach to classification problems with feature space heterogeneity

Kybernetes ◽  
2019 ◽  
Vol 48 (9) ◽  
pp. 2006-2029
Author(s):  
Hongshan Xiao ◽  
Yu Wang
Keyword(s):  
Factor Analysis ◽  
Meta Analysis ◽  
Feature Space ◽  
Classification Performance ◽  
Classification Algorithm ◽  
Significant Feature ◽  
Data Sets ◽  
Data Set ◽  
Classification Techniques ◽  
Content Type

Purpose Feature space heterogeneity exists widely in various application fields of classification techniques, such as customs inspection decision, credit scoring and medical diagnosis. This paper aims to study the relationship between feature space heterogeneity and classification performance. Design/methodology/approach A measurement is first developed for measuring and identifying any significant heterogeneity that exists in the feature space of a data set. The main idea of this measurement is derived from a meta-analysis. For the data set with significant feature space heterogeneity, a classification algorithm based on factor analysis and clustering is proposed to learn the data patterns, which, in turn, are used for data classification. Findings The proposed approach has two main advantages over the previous methods. The first advantage lies in feature transform using orthogonal factor analysis, which results in new features without redundancy and irrelevance. The second advantage rests on samples partitioning to capture the feature space heterogeneity reflected by differences of factor scores. The validity and effectiveness of the proposed approach is verified on a number of benchmarking data sets. Research limitations/implications Measurement should be used to guide the heterogeneity elimination process, which is an interesting topic in future research. In addition, to develop a classification algorithm that enables scalable and incremental learning for large data sets with significant feature space heterogeneity is also an important issue. Practical implications Measuring and eliminating the feature space heterogeneity possibly existing in the data are important for accurate classification. This study provides a systematical approach to feature space heterogeneity measurement and elimination for better classification performance, which is favorable for applications of classification techniques in real-word problems. Originality/value A measurement based on meta-analysis for measuring and identifying any significant feature space heterogeneity in a classification problem is developed, and an ensemble classification framework is proposed to deal with the feature space heterogeneity and improve the classification accuracy.

scholarly journals Deep generative models in DataSHIELD

2021 ◽  
Vol 21 (1) ◽  
Author(s):  
Stefan Lenz ◽  
Moritz Hess ◽  
Harald Binder
Keyword(s):  
Genetic Variant ◽  
Small Sample Size ◽  
Synthetic Data ◽  
Routine Data ◽  
Original Data ◽  
Generative Models ◽  
Small Sample ◽  
Generative Adversarial Networks ◽  
Artificial Data ◽  
Data Set

Abstract Background The best way to calculate statistics from medical data is to use the data of individual patients. In some settings, this data is difficult to obtain due to privacy restrictions. In Germany, for example, it is not possible to pool routine data from different hospitals for research purposes without the consent of the patients. Methods The DataSHIELD software provides an infrastructure and a set of statistical methods for joint, privacy-preserving analyses of distributed data. The contained algorithms are reformulated to work with aggregated data from the participating sites instead of the individual data. If a desired algorithm is not implemented in DataSHIELD or cannot be reformulated in such a way, using artificial data is an alternative. Generating artificial data is possible using so-called generative models, which are able to capture the distribution of given data. Here, we employ deep Boltzmann machines (DBMs) as generative models. For the implementation, we use the package “BoltzmannMachines” from the Julia programming language and wrap it for use with DataSHIELD, which is based on R. Results We present a methodology together with a software implementation that builds on DataSHIELD to create artificial data that preserve complex patterns from distributed individual patient data. Such data sets of artificial patients, which are not linked to real patients, can then be used for joint analyses. As an exemplary application, we conduct a distributed analysis with DBMs on a synthetic data set, which simulates genetic variant data. Patterns from the original data can be recovered in the artificial data using hierarchical clustering of the virtual patients, demonstrating the feasibility of the approach. Additionally, we compare DBMs, variational autoencoders, generative adversarial networks, and multivariate imputation as generative approaches by assessing the utility and disclosure of synthetic data generated from real genetic variant data in a distributed setting with data of a small sample size. Conclusions Our implementation adds to DataSHIELD the ability to generate artificial data that can be used for various analyses, e.g., for pattern recognition with deep learning. This also demonstrates more generally how DataSHIELD can be flexibly extended with advanced algorithms from languages other than R.

Sign in / Sign up

Export Citation Format

Share Document