Structure-preserving deep learning

Over the past few years, deep learning has risen to the foreground as a topic of massive interest, mainly as a result of successes obtained in solving large-scale image processing tasks. There are multiple challenging mathematical problems involved in applying deep learning: most deep learning methods require the solution of hard optimisation problems, and a good understanding of the trade-off between computational effort, amount of data and model complexity is required to successfully design a deep learning approach for a given problem.. A large amount of progress made in deep learning has been based on heuristic explorations, but there is a growing effort to mathematically understand the structure in existing deep learning methods and to systematically design new deep learning methods to preserve certain types of structure in deep learning. In this article, we review a number of these directions: some deep neural networks can be understood as discretisations of dynamical systems, neural networks can be designed to have desirable properties such as invertibility or group equivariance and new algorithmic frameworks based on conformal Hamiltonian systems and Riemannian manifolds to solve the optimisation problems have been proposed. We conclude our review of each of these topics by discussing some open problems that we consider to be interesting directions for future research.

Download Full-text

Unsupervised phenotypic analysis of cellular images with multi-scale convolutional neural networks

10.1101/361410 ◽

2018 ◽

Cited By ~ 5

Author(s):

William J. Godinez ◽

Imtiaz Hossain ◽

Xian Zhang

Keyword(s):

Neural Networks ◽

Deep Learning ◽

Large Scale ◽

Cellular Imaging ◽

Phenotypic Analysis ◽

Learning Methods ◽

Multi Scale ◽

Full Resolution ◽

Benchmark Datasets ◽

Microscopy Images

AbstractLarge-scale cellular imaging and phenotyping is a widely adopted strategy for understanding biological systems and chemical perturbations. Quantitative analysis of cellular images for identifying phenotypic changes is a key challenge within this strategy, and has recently seen promising progress with approaches based on deep neural networks. However, studies so far require either pre-segmented images as input or manual phenotype annotations for training, or both. To address these limitations, we have developed an unsupervised approach that exploits the inherent groupings within cellular imaging datasets to define surrogate classes that are used to train a multi-scale convolutional neural network. The trained network takes as input full-resolution microscopy images, and, without the need for segmentation, yields as output feature vectors that support phenotypic profiling. Benchmarked on two diverse benchmark datasets, the proposed approach yields accurate phenotypic predictions as well as compound potency estimates comparable to the state-of-the-art. More importantly, we show that the approach identifies novel cellular phenotypes not included in the manual annotation nor detected by previous studies.Author summaryCellular microscopy images provide detailed information about how cells respond to genetic or chemical treatments, and have been widely and successfully used in basic research and drug discovery. The recent breakthrough of deep learning methods for natural imaging recognition tasks has triggered the development and application of deep learning methods to cellular images to understand how cells change upon perturbation. Although successful, deep learning studies so far either can only take images of individual cells as input or require human experts to label a large amount of images. In this paper, we present an unsupervised deep learning approach that, without any human annotation, analyzes directly full-resolution microscopy images displaying typically hundreds of cells. We apply the approach to two benchmark datasets, and show that the approach identifies novel visual phenotypes not detected by previous studies.

Download Full-text

Multilabel Text Classification in News Articles Using Long-Term Memory with Word2Vec

Jurnal RESTI (Rekayasa Sistem dan Teknologi Informasi) ◽

10.29207/resti.v4i2.1655 ◽

2020 ◽

Vol 4 (2) ◽

pp. 276-285

Author(s):

Winda Kurnia Sari ◽

Dian Palupi Rini ◽

Reza Firsandaya Malik ◽

Iman Saladin B. Azhar

Keyword(s):

Neural Networks ◽

Deep Learning ◽

Text Classification ◽

Large Scale ◽

Short Term Memory ◽

Short Term ◽

Learning Methods ◽

Processing Variable ◽

Term Memory ◽

Long Short Term Memory

Multilabel text classification is a task of categorizing text into one or more categories. Like other machine learning, multilabel classification performance is limited to the small labeled data and leads to the difficulty of capturing semantic relationships. It requires a multilabel text classification technique that can group four labels from news articles. Deep Learning is a proposed method for solving problems in multilabel text classification techniques. Some of the deep learning methods used for text classification include Convolutional Neural Networks, Autoencoders, Deep Belief Networks, and Recurrent Neural Networks (RNN). RNN is one of the most popular architectures used in natural language processing (NLP) because the recurrent structure is appropriate for processing variable-length text. One of the deep learning methods proposed in this study is RNN with the application of the Long Short-Term Memory (LSTM) architecture. The models are trained based on trial and error experiments using LSTM and 300-dimensional words embedding features with Word2Vec. By tuning the parameters and comparing the eight proposed Long Short-Term Memory (LSTM) models with a large-scale dataset, to show that LSTM with features Word2Vec can achieve good performance in text classification. The results show that text classification using LSTM with Word2Vec obtain the highest accuracy is in the fifth model with 95.38, the average of precision, recall, and F1-score is 95. Also, LSTM with the Word2Vec feature gets graphic results that are close to good-fit on seventh and eighth models.

Download Full-text

Computational Complexity Reduction of Neural Networks of Brain Tumor Image Segmentation by Introducing Fermi–Dirac Correction Functions

Entropy ◽

10.3390/e23020223 ◽

2021 ◽

Vol 23 (2) ◽

pp. 223

Author(s):

Yen-Ling Tai ◽

Shin-Jhe Huang ◽

Chien-Chang Chen ◽

Henry Horng-Shing Lu

Keyword(s):

Neural Networks ◽

Deep Learning ◽

Computational Complexity ◽

High Performance ◽

Low Cost ◽

Structural Complexity ◽

Correction Function ◽

Computational Time ◽

Learning Methods ◽

Band Theory

Nowadays, deep learning methods with high structural complexity and flexibility inevitably lean on the computational capability of the hardware. A platform with high-performance GPUs and large amounts of memory could support neural networks having large numbers of layers and kernels. However, naively pursuing high-cost hardware would probably drag the technical development of deep learning methods. In the article, we thus establish a new preprocessing method to reduce the computational complexity of the neural networks. Inspired by the band theory of solids in physics, we map the image space into a noninteraction physical system isomorphically and then treat image voxels as particle-like clusters. Then, we reconstruct the Fermi–Dirac distribution to be a correction function for the normalization of the voxel intensity and as a filter of insignificant cluster components. The filtered clusters at the circumstance can delineate the morphological heterogeneity of the image voxels. We used the BraTS 2019 datasets and the dimensional fusion U-net for the algorithmic validation, and the proposed Fermi–Dirac correction function exhibited comparable performance to other employed preprocessing methods. By comparing to the conventional z-score normalization function and the Gamma correction function, the proposed algorithm can save at least 38% of computational time cost under a low-cost hardware architecture. Even though the correction function of global histogram equalization has the lowest computational time among the employed correction functions, the proposed Fermi–Dirac correction function exhibits better capabilities of image augmentation and segmentation.

Download Full-text

Classification of Skin Disease Using Deep Learning Neural Networks with MobileNet V2 and LSTM

Sensors ◽

10.3390/s21082852 ◽

2021 ◽

Vol 21 (8) ◽

pp. 2852

Author(s):

Parvathaneni Naga Srinivasu ◽

Jalluri Gnana SivaSai ◽

Muhammad Fazal Ijaz ◽

Akash Kumar Bhoi ◽

Wonjoon Kim ◽

...

Keyword(s):

Neural Network ◽

Neural Networks ◽

Deep Learning ◽

Convolutional Neural Network ◽

Skin Disease ◽

Network Architecture ◽

Large Scale ◽

Short Term Memory ◽

Convolutional Networks ◽

Occurrence Matrix

Deep learning models are efficient in learning the features that assist in understanding complex patterns precisely. This study proposed a computerized process of classifying skin disease through deep learning based MobileNet V2 and Long Short Term Memory (LSTM). The MobileNet V2 model proved to be efficient with a better accuracy that can work on lightweight computational devices. The proposed model is efficient in maintaining stateful information for precise predictions. A grey-level co-occurrence matrix is used for assessing the progress of diseased growth. The performance has been compared against other state-of-the-art models such as Fine-Tuned Neural Networks (FTNN), Convolutional Neural Network (CNN), Very Deep Convolutional Networks for Large-Scale Image Recognition developed by Visual Geometry Group (VGG), and convolutional neural network architecture that expanded with few changes. The HAM10000 dataset is used and the proposed method has outperformed other methods with more than 85% accuracy. Its robustness in recognizing the affected region much faster with almost 2× lesser computations than the conventional MobileNet model results in minimal computational efforts. Furthermore, a mobile application is designed for instant and proper action. It helps the patient and dermatologists identify the type of disease from the affected region’s image at the initial stage of the skin disease. These findings suggest that the proposed system can help general practitioners efficiently and effectively diagnose skin conditions, thereby reducing further complications and morbidity.

Download Full-text

EyeHealer: A large-scale anterior eye segment dataset with eye structure and lesion annotations

Precision Clinical Medicine ◽

10.1093/pcmedi/pbab009 ◽

2021 ◽

Author(s):

Wenjia Cai ◽

Jie Xu ◽

Ke Wang ◽

Xiaohong Liu ◽

Wenqin Xu ◽

...

Keyword(s):

Deep Learning ◽

Large Scale ◽

Image Annotation ◽

Clinical Care ◽

Significant Proportion ◽

Anterior Segment ◽

Future Research ◽

Learning Models ◽

Anterior Eye Segment ◽

Segmentation Models

Abstract Anterior segment eye diseases account for a significant proportion of presentations to eye clinics worldwide, including diseases associated with corneal pathologies, anterior chamber abnormalities (e.g. blood or inflammation) and lens diseases. The construction of an automatic tool for the segmentation of anterior segment eye lesions will greatly improve the efficiency of clinical care. With research on artificial intelligence progressing in recent years, deep learning models have shown their superiority in image classification and segmentation. The training and evaluation of deep learning models should be based on a large amount of data annotated with expertise, however, such data are relatively scarce in the domain of medicine. Herein, the authors developed a new medical image annotation system, called EyeHealer. It is a large-scale anterior eye segment dataset with both eye structures and lesions annotated at the pixel level. Comprehensive experiments were conducted to verify its performance in disease classification and eye lesion segmentation. The results showed that semantic segmentation models outperformed medical segmentation models. This paper describes the establishment of the system for automated classification and segmentation tasks. The dataset will be made publicly available to encourage future research in this area.

Download Full-text

Large-Scale Modeling of Multispecies Acute Toxicity End Points Using Consensus of Multitask Deep Learning Methods

Journal of Chemical Information and Modeling ◽

10.1021/acs.jcim.0c01164 ◽

2021 ◽

Vol 61 (2) ◽

pp. 653-663

Author(s):

Sankalp Jain ◽

Vishal B. Siramshetty ◽

Vinicius M. Alves ◽

Eugene N. Muratov ◽

Nicole Kleinstreuer ◽

...

Keyword(s):

Deep Learning ◽

Acute Toxicity ◽

Large Scale ◽

Learning Methods ◽

End Points ◽

Scale Modeling ◽

Large Scale Modeling

Download Full-text

MetaLearning with Graph Neural Networks

ACM SIGKDD Explorations Newsletter ◽

10.1145/3510374.3510379 ◽

2021 ◽

Vol 23 (2) ◽

pp. 13-22

Author(s):

Debmalya Mandal ◽

Sourav Medya ◽

Brian Uzzi ◽

Charu Aggarwal

Keyword(s):

Neural Networks ◽

Deep Neural Networks ◽

Future Research ◽

Open Problems ◽

Research Directions ◽

Graph Problems ◽

Meta Learning ◽

Comprehensive Survey ◽

Future Research Directions ◽

Graph Neural Networks

Graph Neural Networks (GNNs), a generalization of deep neural networks on graph data have been widely used in various domains, ranging from drug discovery to recommender systems. However, GNNs on such applications are limited when there are few available samples. Meta-learning has been an important framework to address the lack of samples in machine learning, and in recent years, researchers have started to apply meta-learning to GNNs. In this work, we provide a comprehensive survey of different metalearning approaches involving GNNs on various graph problems showing the power of using these two approaches together. We categorize the literature based on proposed architectures, shared representations, and applications. Finally, we discuss several exciting future research directions and open problems.

Download Full-text

Human Skin Detection in Color Images Using Deep Learning

International Journal of Computer Vision and Image Processing ◽

10.4018/ijcvip.2015070101 ◽

2015 ◽

Vol 5 (2) ◽

pp. 1-13 ◽

Cited By ~ 1

Author(s):

Mohammadreza Hajiarbabi ◽

Arvin Agah

Keyword(s):

Neural Networks ◽

Deep Learning ◽

Human Skin ◽

Skin Color ◽

Color Image ◽

Gaussian Model ◽

Color Images ◽

Skin Detection ◽

Learning Methods ◽

Rule Based

Human skin detection is an important and challenging problem in computer vision. Skin detection can be used as the first phase in face detection when using color images. The differences in illumination and ranges of skin colors have made skin detection a challenging task. Gaussian model, rule based methods, and artificial neural networks are methods that have been used for human skin color detection. Deep learning methods are new techniques in learning that have shown improved classification power compared to neural networks. In this paper the authors use deep learning methods in order to enhance the capabilities of skin detection algorithms. Several experiments have been performed using auto encoders and different color spaces. The proposed technique is evaluated compare with other available methods in this domain using two color image databases. The results show that skin detection utilizing deep learning has better results compared to other methods such as rule-based, Gaussian model and feed forward neural network.

Download Full-text

A Comprehensive Taxonomy of Dynamic Texture Representation

ACM Computing Surveys ◽

10.1145/3487892 ◽

2023 ◽

Vol 55 (1) ◽

pp. 1-39

Author(s):

Thanh Tuan Nguyen ◽

Thanh Phuong Nguyen

Keyword(s):

Large Scale ◽

Environmental Changes ◽

State Of The Art ◽

The State ◽

Future Research ◽

Research Activities ◽

Potential Applications ◽

Benchmark Datasets ◽

Negative Impacts ◽

Made In

Representing dynamic textures (DTs) plays an important role in many real implementations in the computer vision community. Due to the turbulent and non-directional motions of DTs along with the negative impacts of different factors (e.g., environmental changes, noise, illumination, etc.), efficiently analyzing DTs has raised considerable challenges for the state-of-the-art approaches. For 20 years, many different techniques have been introduced to handle the above well-known issues for enhancing the performance. Those methods have shown valuable contributions, but the problems have been incompletely dealt with, particularly recognizing DTs on large-scale datasets. In this article, we present a comprehensive taxonomy of DT representation in order to purposefully give a thorough overview of the existing methods along with overall evaluations of their obtained performances. Accordingly, we arrange the methods into six canonical categories. Each of them is then taken in a brief presentation of its principal methodology stream and various related variants. The effectiveness levels of the state-of-the-art methods are then investigated and thoroughly discussed with respect to quantitative and qualitative evaluations in classifying DTs on benchmark datasets. Finally, we point out several potential applications and the remaining challenges that should be addressed in further directions. In comparison with two existing shallow DT surveys (i.e., the first one is out of date as it was made in 2005, while the newer one (published in 2016) is an inadequate overview), we believe that our proposed comprehensive taxonomy not only provides a better view of DT representation for the target readers but also stimulates future research activities.

Download Full-text

An Investigation of Deep Learning Models for EEG-Based Emotion Recognition

Frontiers in Neuroscience ◽

10.3389/fnins.2020.622759 ◽

2020 ◽

Vol 14 ◽

Author(s):

Yaqing Zhang ◽

Jinling Chen ◽

Jen Hong Tan ◽

Yuxuan Chen ◽

Yunyi Chen ◽

...

Keyword(s):

Machine Learning ◽

Neural Networks ◽

Deep Learning ◽

Emotion Recognition ◽

Real Life ◽

Learning Rate ◽

Learning Models ◽

Learning Methods ◽

Time Frequency ◽

Comparison Results

Emotion is the human brain reacting to objective things. In real life, human emotions are complex and changeable, so research into emotion recognition is of great significance in real life applications. Recently, many deep learning and machine learning methods have been widely applied in emotion recognition based on EEG signals. However, the traditional machine learning method has a major disadvantage in that the feature extraction process is usually cumbersome, which relies heavily on human experts. Then, end-to-end deep learning methods emerged as an effective method to address this disadvantage with the help of raw signal features and time-frequency spectrums. Here, we investigated the application of several deep learning models to the research field of EEG-based emotion recognition, including deep neural networks (DNN), convolutional neural networks (CNN), long short-term memory (LSTM), and a hybrid model of CNN and LSTM (CNN-LSTM). The experiments were carried on the well-known DEAP dataset. Experimental results show that the CNN and CNN-LSTM models had high classification performance in EEG-based emotion recognition, and their accurate extraction rate of RAW data reached 90.12 and 94.17%, respectively. The performance of the DNN model was not as accurate as other models, but the training speed was fast. The LSTM model was not as stable as the CNN and CNN-LSTM models. Moreover, with the same number of parameters, the training speed of the LSTM was much slower and it was difficult to achieve convergence. Additional parameter comparison experiments with other models, including epoch, learning rate, and dropout probability, were also conducted in the paper. Comparison results prove that the DNN model converged to optimal with fewer epochs and a higher learning rate. In contrast, the CNN model needed more epochs to learn. As for dropout probability, reducing the parameters by ~50% each time was appropriate.

Download Full-text