Search results for: convolutional coding

Authors: Chengcao Sun, Shujun Li, Cuili Yang, Yongyong Xi, Liang Wang, Feng Zhang, Dejia Li

Abstract:

Recently, the long non-coding RNA (lncRNA) NEAT1 has been identified as an oncogenic gene in multiple cancer types and elevated expression of NEAT1 was tightly linked to tumorigenesis and cancer progression. However, the molecular basis for this observation has not been characterized in progression of non-small cell lung cancer (NSCLC). In our studies, we identified NEAT1 was highly expressed in NSCLC patients and was a novel regulator of NSCLC progression. Patients whose tumors had high NEAT1 expression had a shorter overall survival than patients whose tumors had low NEAT1 expression. Further, NEAT1 significantly accelerates NSCLC cell growth and metastasis in vitro and tumor growth in vivo. Additionally, by using bioinformatics study and RNA pull down combined with luciferase reporter assays, we demonstrated that NEAT1 functioned as a competing endogenous RNA (ceRNA) for has-miR-377-3p, antagonized its functions and led to the de-repression of its endogenous targets E2F3, which was a core oncogene in promoting NSCLC progression. Taken together, these observations imply that the NEAT1 modulated the expression of E2F3 gene by acting as a competing endogenous RNA, which may build up the missing link between the regulatory miRNA network and NSCLC progression.

Keywords: long non-coding RNA NEAT1, hsa-miRNA-377-3p, E2F3, non-small cell lung cancer, tumorigenesis

Procedia PDF Downloads 342

Authors: Farina Riaz, Shahab Abdulla, Srinjoy Ganguly, Hajime Suzuki, Ravinesh C. Deo, Susan Hopkins

Abstract:

Recently, quantum machine learning has received significant attention. For various types of data, including text and images, numerous quantum machine learning (QML) models have been created and are being tested. Images are exceedingly complex data components that demand more processing power. Despite being mature, classical machine learning still has difficulties with big data applications. Furthermore, quantum technology has revolutionized how machine learning is thought of, by employing quantum features to address optimization issues. Since quantum hardware is currently extremely noisy, it is not practicable to run machine learning algorithms on it without risking the production of inaccurate results. To discover the advantages of quantum versus classical approaches, this research has concentrated on colored image data. Deep learning classification models are currently being created on Quantum platforms, but they are still in a very early stage. Black and white benchmark image datasets like MNIST and Fashion MINIST have been used in recent research. MNIST and CIFAR-10 were compared for binary classification, but the comparison showed that MNIST performed more accurately than colored CIFAR-10. This research will evaluate the performance of the QML algorithm on the colored benchmark dataset CIFAR-10 to advance QML's real-time applicability. However, deep learning classification models have not been developed to compare colored images like Quantum Convolutional Neural Network (QCNN) to determine how much it is better to classical. Only a few models, such as quantum variational circuits, take colored images. The methodology adopted in this research is a hybrid approach by using penny lane as a simulator. To process the 10 classes of CIFAR-10, the image data has been translated into grey scale and the 28 × 28-pixel image containing 10,000 test and 50,000 training images were used. The objective of this work is to determine how much the quantum approach can outperform a classical approach for a comprehensive dataset of color images. After pre-processing 50,000 images from a classical computer, the QCNN model adopted a hybrid method and encoded the images into a quantum simulator for feature extraction using quantum gate rotations. The measurements were carried out on the classical computer after the rotations were applied. According to the results, we note that the QCNN approach is ~12% more effective than the traditional classical CNN approaches and it is possible that applying data augmentation may increase the accuracy. This study has demonstrated that quantum machine and deep learning models can be relatively superior to the classical machine learning approaches in terms of their processing speed and accuracy when used to perform classification on colored classes.

Keywords: CIFAR-10, quantum convolutional neural networks, quantum deep learning, quantum machine learning

Procedia PDF Downloads 90

Authors: Lamis Naddaf, Yuval Tabach

Abstract:

In this research, we identified the missense genetic variants that have the potential to enhance resistance against cancer. Such field has not been widely explored, as researchers tend to investigate mutations that cause diseases, in response to the suffering of patients, rather than those mutations that protect from them. In conjunction with the genomic revolution, and the advances in genetic engineering and synthetic biology, identifying the protective variants will increase the power of genotype-phenotype predictions and can have significant implications on improved risk estimation, diagnostics, prognosis and even for personalized therapy and drug discovery. To approach our goal, we systematically investigated the sites of the coding genomes and picked up the alleles that showed a correlation with the species’ cancer resistance. We predicted 250 protecting variants (PVs) with a 0.01 false discovery rate and more than 20 thousand PVs with a 0.25 false discovery rate. Cancer resistance in Mammals and reptiles was significantly predicted by the number of PVs a species has. Moreover, Genes enriched with the protecting variants are enriched in pathways relevant to tumor suppression like pathways of Hedgehog signaling and silencing, which its improper activation is associated with the most common form of cancer malignancy. We also showed that the PVs are more abundant in healthy people compared to cancer patients within different human races.

Keywords: comparative genomics, machine learning, cancer resistance, cancer-protecting alleles

Procedia PDF Downloads 69

Authors: Lamis Naddaf, Yuval Tabach

Abstract:

We identified missense genetic variants with the potential to enhance resistance against cancer. Such a field has not been widely explored as researchers tend to investigate the mutations that cause diseases, in response to the suffering of patients, rather than those mutations that protect from them. In conjunction with the genomic revolution and the advances in genetic engineering and synthetic biology, identifying the protective variants will increase the power of genotype-phenotype predictions and have significant implications for improved risk estimation, diagnostics, prognosis, and even personalized therapy and drug discovery. To approach our goal, we systematically investigated the sites of the coding genomes and selected the alleles that showed a correlation with the species’ cancer resistance. Interestingly, we found several amino acids that are more generally preferred (like the Proline) or avoided (like the Cysteine) by the resistant species. Furthermore, Cancer resistance in mammals and reptiles is significantly predicted by the number of the predicted protecting variants (PVs) a species has. Moreover, PVs-enriched-genes are enriched in pathways relevant to tumor suppression. For example, they are enriched in the Hedgehog signaling and silencing pathways, which its improper activation is associated with the most common form of cancer malignancy. We also showed that the PVs are mostly more abundant in healthy people compared to cancer patients within different human races.

Keywords: cancer resistance, protecting variant, naked mole rat, comparative genomics

Procedia PDF Downloads 71

Authors: Diana Mariz Catangay, Irish-Ann Montano, Frances Janine Suyat

Abstract:

As an active pedestal for student’s interaction with both issues happening inside the school and out; may it be political, societal, international, or other current events, a school paper should at least meet the standard of providing a representation of the school’s morals and values and help the institution uplift its image. The researchers seek to ascertain how the two student publications from the Philippines’ two prime state universities, the University of the Philippines’ Philippine Collegian, and the Polytechnic University of the Philippines’ the Catalyst, presents iii their school through balanced journalism and objective documentation. The objectives include determining the number of school-related articles published versus those articles that are concerned outside the school’s jurisdiction, analyzing the insight it provides on the image of the university, assessing the similarities and/or differences between the two publications, and, finally, coming up with the conclusion of how the two newspapers uses their medium to present their respective schools. The research used the quantitative method of research in order to further analyze the articles that will serve as bases in coming up with the right conclusion based on the objectives of the study. Coding sheets and coding guides are utilized for the chosen research method. The gathered findings will then be interpreted as fitting to the goal of the research.

Keywords: content analysis, journalistic presentation, student publications, state universities

Procedia PDF Downloads 155

Authors: Mehrnaz Mostafavi

Abstract:

Recent advances in single-molecule protein identification (ID) and quantification techniques are poised to revolutionize proteomics, enabling researchers to delve into single-cell proteomics and identify low-abundance proteins crucial for biomedical and clinical research. This paper introduces a different approach to single-molecule protein ID and quantification using tri-color amino acid tags and a plasmonic nanopore device. A comprehensive simulator incorporating various physical phenomena was designed to predict and model the device's behavior under diverse experimental conditions, providing insights into its feasibility and limitations. The study employs a whole-proteome single-molecule identification algorithm based on convolutional neural networks, achieving high accuracies (>90%), particularly in challenging conditions (95–97%). To address potential challenges in clinical samples, where post-translational modifications affecting labeling efficiency, the paper evaluates protein identification accuracy under partial labeling conditions. Solid-state nanopores, capable of processing tens of individual proteins per second, are explored as a platform for this method. Unlike techniques relying solely on ion-current measurements, this approach enables parallel readout using high-density nanopore arrays and multi-pixel single-photon sensors. Convolutional neural networks contribute to the method's versatility and robustness, simplifying calibration procedures and potentially allowing protein ID based on partial reads. The study also discusses the efficacy of the approach in real experimental conditions, resolving functionally similar proteins. The theoretical analysis, protein labeler program, finite difference time domain calculation of plasmonic fields, and simulation of nanopore-based optical sensing are detailed in the methods section. The study anticipates further exploration of temporal distributions of protein translocation dwell-times and the impact on convolutional neural network identification accuracy. Overall, the research presents a promising avenue for advancing single-molecule protein identification and quantification with broad applications in proteomics research. The contributions made in methodology, accuracy, robustness, and technological exploration collectively position this work at the forefront of transformative developments in the field.

Keywords: nano proteomics, nanopore-based optical sensing, deep learning, artificial intelligence

Procedia PDF Downloads 35

Authors: Leonard Lipovich, Pattaraporn Thepsuwan, Anton-Scott Goustin, Juan Cai, Donghong Ju, James B. Brown

Abstract:

Two-thirds of human genes do not encode any known proteins. Aside from long non-coding RNA (lncRNA) genes with recently-discovered functions, the ~40,000 non-protein-coding human genes remain poorly understood, and a role for their transcripts as de-facto unconventional messenger RNAs has not been formally excluded. Ribosome profiling (Riboseq) predicts translational potential, but without independent evidence of proteins from lncRNA open reading frames (ORFs), ribosome binding of lncRNAs does not prove translation. Previously, we mass-spectrometrically documented translation of specific lncRNAs in human K562 and GM12878 cells. We now examined lncRNA translation in human MCF7 cells, integrating strand-specific Illumina RNAseq, Riboseq, and deep mass spectrometry in biological quadruplicates performed at two core facilities (BGI, China; City of Hope, USA). We excluded known-protein matches. UCSC Genome Browser-assisted manual annotation of imperfect (tryptic-digest-peptides)-to-(lncRNA-three-frame-translations) alignments revealed three peptides hypothetically explicable by 'stop-to-nonstop' in-frame replacement of stop codons by amino acids in two ORFs of the lncRNA MMP24-AS1. To search for this phenomenon genomewide, we designed and implemented a novel pipeline, matching tryptic-digest spectra to wildcard-instead-of-stop versions of repeat-masked, six-frame, whole-genome translations. Along with singleton putative stop-to-nonstop events affecting four other lncRNAs, we identified 24 additional peptides with stop-to-nonstop in-frame substitutions from multiple positive-strand MMP24-AS1 ORFs. Only UAG and UGA, never UAA, stop codons were impacted. All MMP24-AS1-matching spectra met the same significance thresholds as high-confidence known-protein signatures. Targeted resequencing of MMP24-AS1 genomic DNA and cDNA from the same samples did not reveal any mutations, polymorphisms, or sequencing-detectable RNA editing. This unprecedented apparent gene-specific violation of the genetic code highlights the importance of matching peptides to whole-genome, not known-genes-only, ORFs in mass-spectrometry workflows, and suggests a new mechanism enhancing the combinatorial complexity of the proteome. Funding: NIH Director’s New Innovator Award 1DP2-CA196375 to LL.

Keywords: genetic code, lncRNA, long non-coding RNA, mass spectrometry, proteogenomics, ribo-seq, ribosome, RNAseq

Procedia PDF Downloads 201

Authors: T. C. C. Soo, S. Bhassu

Abstract:

Unlike the common presence of both innate and adaptive immunity in vertebrates, crustaceans, in particular, shrimps, have been discovered to possess only innate immunity. This further emphasizes the importance of innate immunity within shrimps in pathogenic resistance. Under the study of pathogenic immune challenge, different shrimp species actually exhibit varying degrees of immune resistance towards the same pathogen. Furthermore, even within the same shrimp species, different batches of challenged shrimps can have different strengths of immune defence. Several important pathways are activated within shrimps during pathogenic infection. One of them is JAK-STAT pathway that is activated during bacterial, viral and fungal infections by which STAT(Signal Transducer and Activator of Transcription) gene is the core element of the pathway. Based on theory of Central Dogma, the genomic information is transmitted in the order of DNA, RNA and protein. This study is focused in uncovering the important evolutionary patterns present within the DNA (non-coding region) and RNA (coding region). The three shrimp species involved are Macrobrachium rosenbergii, Penaeus monodon and Litopenaeus vannamei which all possess commercial significance. The shrimp species were challenged with a famous penaeid shrimp virus called white spot syndrome virus (WSSV) which can cause serious lethality. Tissue samples were collected during time intervals of 0h, 3h, 6h, 12h, 24h, 36h and 48h. The DNA and RNA samples were then extracted using conventional kits from the hepatopancreas tissue samples. PCR technique together with designed STAT gene conserved primers were utilized for identification of the STAT coding sequences using RNA-converted cDNA samples and subsequent characterization using various bioinformatics approaches including Ramachandran plot, ProtParam and SWISS-MODEL. The varying levels of immune STAT gene activation for the three shrimp species during WSSV infection were confirmed using qRT-PCR technique. For one sample, three biological replicates with three technical replicates each were used for qRT-PCR. On the other hand, DNA samples were important for uncovering the structural variations within the genomic region of STAT gene which would greatly assist in understanding the STAT protein functional variations. The partially-overlapping primers technique was used for the genomic region sequencing. The evolutionary inferences and event predictions were then conducted through the Bayesian Inference method using all the acquired coding and non-coding sequences. This was supplemented by the construction of conventional phylogenetic trees using Maximum likelihood method. The results showed that adaptive evolution caused STAT gene sequence mutations between different shrimp species which led to evolutionary divergence event. Subsequently, the divergent sites were correlated to the differing expressions of STAT gene. Ultimately, this study assists in knowing the shrimp species innate immune variability and selection of disease resistant shrimps for breeding purpose. The deeper understanding of STAT gene evolution from the perspective of both purifying and adaptive approaches not only can provide better immunological insight among shrimp species, but also can be used as a good reference for immunological studies in humans or other model organisms.

Keywords: gene evolution, JAK-STAT pathway, immunology, STAT gene

Procedia PDF Downloads 116

Authors: A. Chillali, M. Sahmoudi

Abstract:

In this paper we will give a method for encoding the elements of the ring of integers of sextic extension, namely L = Q(a,b) which is a rational quadratic over cubic field K =Q(a) where a^{2} is a rational square free integer and b is a root of irreducible polynomiale of degree 3.

Keywords: coding, integral bases, sextic, quadratic

Procedia PDF Downloads 262

Authors: Yuanchao Sang, Zhihao Gong, Longsheng Chen, Long Chen

Abstract:

Clothing fashion represents a human’s aesthetic appreciation towards everyday outfits and appetite for fashion, and it reflects the development of status in society, humanity, and economics. However, modelling fashion by machine is extremely challenging because fashion is too abstract to be efficiently described by machines. Even human beings can hardly reach a consensus about fashion. In this paper, we are dedicated to answering a fundamental fashion-related problem: what image feature best describes clothing fashion? To address this issue, we have designed and evaluated various image features, ranging from traditional low-level hand-crafted features to mid-level style awareness features to various current popular deep neural network-based features, which have shown state-of-the-art performance in various vision tasks. In summary, we tested the following 9 feature representations: color, texture, shape, style, convolutional neural networks (CNNs), CNNs with distance metric learning (CNNs&DML), AutoEncoder, CNNs with multiple layer combination (CNNs&MLC) and CNNs with dynamic feature clustering (CNNs&DFC). Finally, we validated the performance of these features on two publicly available datasets. Quantitative and qualitative experimental results on both intra-domain and inter-domain fashion clothing image retrieval showed that deep learning based feature representations far outweigh traditional hand-crafted feature representation. Additionally, among all deep learning based methods, CNNs with explicit feature clustering performs best, which shows feature clustering is essential for discriminative fashion feature representation.

Keywords: convolutional neural network, feature representation, image processing, machine modelling

Procedia PDF Downloads 109

Authors: Bahareh Golchin, Nooshin Riahi

Abstract:

One of the most significant issues as attended a lot in recent years is that of recognizing the sentiments and emotions in social media texts. The analysis of sentiments and emotions is intended to recognize the conceptual information such as the opinions, feelings, attitudes and emotions of people towards the products, services, organizations, people, topics, events and features in the written text. These indicate the greatness of the problem space. In the real world, businesses and organizations are always looking for tools to gather ideas, emotions, and directions of people about their products, services, or events related to their own. This article uses the Twitter social network, one of the most popular social networks with about 420 million active users, to extract data. Using this social network, users can share their information and opinions about personal issues, policies, products, events, etc. It can be used with appropriate classification of emotional states due to the availability of its data. In this study, supervised learning and deep neural network algorithms are used to classify the emotional states of Twitter users. The use of deep learning methods to increase the learning capacity of the model is an advantage due to the large amount of available data. Tweets collected on various topics are classified into four classes using a combination of two Bidirectional Long Short Term Memory network and a Convolutional network. The results obtained from this study with an average accuracy of 93%, show good results extracted from the proposed framework and improved accuracy compared to previous work.

Keywords: emotion classification, sentiment analysis, social networks, deep neural networks

Procedia PDF Downloads 107

Authors: F. Ghazalnaz Sharifonnasabi, Iman Makhdoom

Abstract:

Breast cancer is a serious health concern that affects many people around the world. According to a study published in the Breast journal, the global burden of breast cancer is expected to increase significantly over the next few decades. The number of deaths from breast cancer has been increasing over the years, but the age-standardized mortality rate has decreased in some countries. It’s important to be aware of the risk factors for breast cancer and to get regular check- ups to catch it early if it does occur. Machin learning techniques have been used to aid in the early detection and diagnosis of breast cancer. These techniques, that have been shown to be effective in predicting and diagnosing the disease, have become a research hotspot. In this study, we consider two deep learning approaches including: Multi-Layer Perceptron (MLP), and Convolutional Neural Network (CNN). We also considered the five-machine learning algorithm titled: Decision Tree (C4.5), Naïve Bayesian (NB), Support Vector Machine (SVM), K-Nearest Neighbors (KNN) Algorithm and XGBoost (eXtreme Gradient Boosting) on the Breast Cancer Wisconsin Diagnostic dataset. We have carried out the process of evaluating and comparing classifiers involving selecting appropriate metrics to evaluate classifier performance and selecting an appropriate tool to quantify this performance. The main purpose of the study is predicting and diagnosis breast cancer, applying the mentioned algorithms and also discovering of the most effective with respect to confusion matrix, accuracy and precision. It is realized that CNN outperformed all other classifiers and achieved the highest accuracy (0.982456). The work is implemented in the Anaconda environment based on Python programing language.

Keywords: breast cancer, multi-layer perceptron, Naïve Bayesian, SVM, decision tree, convolutional neural network, XGBoost, KNN

Procedia PDF Downloads 41

Authors: Anastasia Sorokina

Abstract:

Graduate language loss or attrition has been well documented in individuals who migrate and become emersed in a different language environment. This phenomenon of first language (L1) attrition is an example of non-pathological (not due to trauma) and can manifest itself in frequent pauses, search for words, or grammatical errors. While the widely experienced loss of one’s first language might seem harmless, there is convincing evidence from the disciplines of Developmental Psychology, Bilingual Studies, and even Psychotherapy that language plays a crucial role in the memory of self. In fact, we remember, store, and share personal memories with the help of language. Dual-Coding Theory suggests that language memory code deterioration could lead to forgetting. Yet, no one has investigated a possible connection between language loss and memory. The present study aims to address this research gap by examining a corpus of 1,495 memories of Russian-English bilinguals who are on a continuum of L1 (first language) attrition. Since phenomenological properties capture how well a memory is remembered, the following descriptors were selected - vividness, ease of recall, emotional valence, personal significance, and confidence in the event. A series of linear regression statistical analyses were run to examine the possible negative effects of L1 attrition on autobiographical memory. The results revealed that L1 attrition might compromise perceived vividness and confidence in the event, which is indicative of memory deterioration. These findings suggest the importance of heritage language maintenance in immigrant communities who might be forced to assimilate as language loss might negatively affect the memory of self.

Keywords: L1 attrition, autobiographical memory, language loss, memory phenomenology, dual coding

Procedia PDF Downloads 77

Authors: Oumaima Khlifati, Khadija Baba

Abstract:

Pavement distress is the main factor responsible for the deterioration of road structure durability, damage vehicles, and driver comfort. Transportation agencies spend a high proportion of their funds on pavement monitoring and maintenance. The auscultation of pavement distress was based on the manual survey, which was extremely time consuming, labor intensive, and required domain expertise. Therefore, the automatic distress detection is needed to reduce the cost of manual inspection and avoid more serious damage by implementing the appropriate remediation actions at the right time. Inspired by recent deep learning applications, this paper proposes an algorithm for automatic road distress detection and classification using on the Deep Convolutional Neural Network (DCNN). In this study, the types of pavement distress are classified as transverse or longitudinal cracking, alligator, pothole, and intact pavement. The dataset used in this work is composed of public asphalt pavement images. In order to learn the structure of the different type of distress, the DCNN models are trained and tested as a multi-label classification task. In addition, to get the highest accuracy for our model, we adjust the structural optimization hyper parameters such as the number of convolutions and max pooling, filers, size of filters, loss functions, activation functions, and optimizer and fine-tuning hyper parameters that conclude batch size and learning rate. The optimization of the model is executed by checking all feasible combinations and selecting the best performing one. The model, after being optimized, performance metrics is calculated, which describe the training and validation accuracies, precision, recall, and F1 score.

Keywords: distress pavement, hyperparameters, automatic classification, deep learning

Procedia PDF Downloads 56

Authors: Nkechinyere Joy Olawuyi, Babajide Samuel Afolabi, Bola Ibitoye

Abstract:

This study collected a preprocessed dataset of chest radiographs and formulated a deep neural network model for detecting abnormalities. It also evaluated the performance of the formulated model and implemented a prototype of the formulated model. This was with the view to developing a deep neural network model to automatically classify abnormalities in chest radiographs. In order to achieve the overall purpose of this research, a large set of chest x-ray images were sourced for and collected from the CheXpert dataset, which is an online repository of annotated chest radiographs compiled by the Machine Learning Research Group, Stanford University. The chest radiographs were preprocessed into a format that can be fed into a deep neural network. The preprocessing techniques used were standardization and normalization. The classification problem was formulated as a multi-label binary classification model, which used convolutional neural network architecture to make a decision on whether an abnormality was present or not in the chest radiographs. The classification model was evaluated using specificity, sensitivity, and Area Under Curve (AUC) score as the parameter. A prototype of the classification model was implemented using Keras Open source deep learning framework in Python Programming Language. The AUC ROC curve of the model was able to classify Atelestasis, Support devices, Pleural effusion, Pneumonia, A normal CXR (no finding), Pneumothorax, and Consolidation. However, Lung opacity and Cardiomegaly had a probability of less than 0.5 and thus were classified as absent. Precision, recall, and F1 score values were 0.78; this implies that the number of False Positive and False Negative is the same, revealing some measure of label imbalance in the dataset. The study concluded that the developed model is sufficient to classify abnormalities present in chest radiographs into present or absent.

Keywords: transfer learning, convolutional neural network, radiograph, classification, multi-label

Procedia PDF Downloads 82

Authors: Jaiden Xuan Schraut, Leon Liu, Yiqiao Yin

Abstract:

Computer vision in medical diagnosis has achieved a high level of success in diagnosing diseases with high accuracy. However, conventional classifiers that produce an image to-label result provides insufficient information for medical professionals to judge and raise concerns over the trust and reliability of a model with results that cannot be explained. In order to gain local insight into cancerous regions, separate tasks such as imaging segmentation need to be implemented to aid the doctors in treating patients, which doubles the training time and costs which renders the diagnosis system inefficient and difficult to be accepted by the public. To tackle this issue and drive AI-first medical solutions further, this paper proposes a multi-output network that follows a U-Net architecture for image segmentation output and features an additional convolutional neural networks (CNN) module for auxiliary classification output. Class activation maps are a method of providing insight into a convolutional neural network’s feature maps that leads to its classification but in the case of lung diseases, the region of interest is enhanced by U-net-assisted Class Activation Map (CAM) visualization. Therefore, our proposed model combines image segmentation models and classifiers to crop out only the lung region of a chest X-ray’s class activation map to provide a visualization that improves the explainability and is able to generate classification results simultaneously which builds trust for AI-led diagnosis systems. The proposed U-Net model achieves 97.61% accuracy and a dice coefficient of 0.97 on testing data from the COVID-QU-Ex Dataset which includes both diseased and healthy lungs.

Keywords: multi-output network model, U-net, class activation map, image classification, medical imaging analysis

Procedia PDF Downloads 158

Authors: Sangita Pokhrel, Nalinda Somasiri, Rebecca Jeyavadhanam, Swathi Ganesan

Abstract:

Tourism is a booming industry with huge future potential for global wealth and employment. There are countless data generated over social media sites every day, creating numerous opportunities to bring more insights to decision-makers. The integration of Big Data Technology into the tourism industry will allow companies to conclude where their customers have been and what they like. This information can then be used by businesses, such as those in charge of managing visitor centers or hotels, etc., and the tourist can get a clear idea of places before visiting. The technical perspective of natural language is processed by analysing the sentiment features of online reviews from tourists, and we then supply an enhanced long short-term memory (LSTM) framework for sentiment feature extraction of travel reviews. We have constructed a web review database using a crawler and web scraping technique for experimental validation to evaluate the effectiveness of our methodology. The text form of sentences was first classified through Vader and Roberta model to get the polarity of the reviews. In this paper, we have conducted study methods for feature extraction, such as Count Vectorization and TFIDF Vectorization, and implemented Convolutional Neural Network (CNN) classifier algorithm for the sentiment analysis to decide the tourist’s attitude towards the destinations is positive, negative, or simply neutral based on the review text that they posted online. The results demonstrated that from the CNN algorithm, after pre-processing and cleaning the dataset, we received an accuracy of 96.12% for the positive and negative sentiment analysis.

Keywords: counter vectorization, convolutional neural network, crawler, data technology, long short-term memory, web scraping, sentiment analysis

Procedia PDF Downloads 52

Authors: Linyu Wang, Mingjun Zhu, Jianhong Xiang, Hanyu Jiang

Abstract:

Aiming at the problem that the spectral efficiency of hybrid precoding (HP) is too low in the current millimeter wave (mmWave) massive multiple input multiple output (MIMO) system, this paper proposes a digital joint equivalent channel hybrid precoding algorithm, which is based on the introduction of digital encoding matrix iteration. First, the objective function is expanded to obtain the relation equation, and the pseudo-inverse iterative function of the analog encoder is derived by using the pseudo-inverse method, which solves the problem of greatly increasing the amount of computation caused by the lack of rank of the digital encoding matrix and reduces the overall complexity of hybrid precoding. Secondly, the analog coding matrix and the millimeter-wave sparse channel matrix are combined into an equivalent channel, and then the equivalent channel is subjected to Singular Value Decomposition (SVD) to obtain a digital coding matrix, and then the derived pseudo-inverse iterative function is used to iteratively regenerate the simulated encoding matrix. The simulation results show that the proposed algorithm improves the system spectral efficiency by 10~20%compared with other algorithms and the stability is also improved.

Keywords: mmWave, massive MIMO, hybrid precoding, singular value decompositing, equivalent channel

Procedia PDF Downloads 64

Authors: Hien Thi Dieu Truong, Mahmoud Al-Sarayreh, Pullanagari Reddy, Marlon M. Reis, Richard Archer

Abstract:

New Zealand mānuka honey is a honeybee product derived mainly from Leptospermum scoparium nectar. The potent antibacterial activity of mānuka honey derives principally from methylglyoxal (MGO), in addition to the hydrogen peroxide and other lesser activities present in all honey. MGO is formed from dihydroxyacetone (DHA) unique to L. scoparium nectar. Mānuka honey also has an idiosyncratic phenolic profile that is useful as a chemical maker. Authentic mānuka honey is highly valuable, but almost all honey is formed from natural mixtures of nectars harvested by a hive over a time period. Once diluted by other nectars, mānuka honey irrevocably loses value. We aimed to apply hyperspectral imaging to honey frames before bulk extraction to minimise the dilution of genuine mānuka by other honey and ensure authenticity at the source. This technology is non-destructive and suitable for an industrial setting. Chemometrics using linear Partial Least Squares (PLS) and Support Vector Machine (SVM) showed limited efficacy in interpreting chemical footprints due to large non-linear relationships between predictor and predictand in a large sample set, likely due to honey quality variability across geographic regions. Therefore, an advanced modelling approach, one-dimensional convolutional neural networks (1D-CNN), was investigated for analysing hyperspectral data for extraction of biochemical information from honey. The 1D-CNN model showed superior prediction of honey quality (R² = 0.73, RMSE = 2.346, RPD= 2.56) to PLS (R² = 0.66, RMSE = 2.607, RPD= 1.91) and SVM (R² = 0.67, RMSE = 2.559, RPD=1.98). Classification of mono-floral manuka honey from multi-floral and non-manuka honey exceeded 90% accuracy for all models tried. Overall, this study reveals the potential of HSI and deep learning modelling for automating the evaluation of honey quality in frames.

Keywords: mānuka honey, quality, purity, potency, deep learning, 1D-CNN, chemometrics

Procedia PDF Downloads 101

Authors: Michael Moeti, Khuliso Sigama, Thapelo Samuel Matlala

Abstract:

One of the principal challenges that the world is confronted with is insecurity. The crime rate is increasing exponentially, and protecting our physical assets especially in the motorist industry, is becoming impossible when applying our own strength. The need to develop technological solutions that detect and report theft without any human interference is inevitable. This is critical, especially for vehicle owners, to ensure theft detection and speedy identification towards recovery efforts in cases where a vehicle is missing or attempted theft is taking place. The vehicle theft detection system uses Convolutional Neural Network (CNN) to recognize the driver's face captured using an installed mobile phone device. The location identification function uses a Global Positioning System (GPS) to determine the real-time location of the vehicle. Upon identification of the location, Global System for Mobile Communications (GSM) technology is used to report or notify the vehicle owner about the whereabouts of the vehicle. The installed mobile app was implemented by making use of python as it is undoubtedly the best choice in machine learning. It allows easy access to machine learning algorithms through its widely developed library ecosystem. The graphical user interface was developed by making use of JAVA as it is better suited for mobile development. Google's online database (Firebase) was used as a means of storage for the application. The system integration test was performed using a simple percentage analysis. Sixty (60) vehicle owners participated in this study as a sample, and questionnaires were used in order to establish the acceptability of the system developed. The result indicates the efficiency of the proposed system, and consequently, the paper proposes the use of the system can effectively monitor the vehicle at any given place, even if it is driven outside its normal jurisdiction. More so, the system can be used as a database to detect, locate and report missing vehicles to different security agencies.

Keywords: CNN, location identification, tracking, GPS, GSM

Procedia PDF Downloads 123

Authors: Stephen L. Green, Alexander N. Gorban, Ivan Y. Tyukin

Abstract:

Within the past decade, using Convolutional Neural Networks (CNN)’s to create Deep Learning systems capable of translating Sign Language into text has been a breakthrough in breaking the communication barrier for deaf-mute people. Conventional research on this subject has been concerned with training the network to recognize the fingerspelling gestures of a given language and produce their corresponding alphanumerics. One of the problems with the current developing technology is that images are scarce, with little variations in the gestures being presented to the recognition program, often skewed towards single skin tones and hand sizes that makes a percentage of the population’s fingerspelling harder to detect. Along with this, current gesture detection programs are only trained on one finger spelling language despite there being one hundred and forty-two known variants so far. All of this presents a limitation for traditional exploitation for the state of current technologies such as CNN’s, due to their large number of required parameters. This work aims to present a technology that aims to resolve this issue by combining a pretrained legacy AI system for a generic object recognition task with a corrector method to uptrain the legacy network. This is a computationally efficient procedure that does not require large volumes of data even when covering a broad range of sign languages such as American Sign Language, British Sign Language and Chinese Sign Language (Pinyin). Implementing recent results on method concentration, namely the stochastic separation theorem, an AI system is supposed as an operate mapping an input present in the set of images u ∈ U to an output that exists in a set of predicted class labels q ∈ Q of the alphanumeric that q represents and the language it comes from. These inputs and outputs, along with the interval variables z ∈ Z represent the system’s current state which implies a mapping that assigns an element x ∈ ℝⁿ to the triple (u, z, q). As all xi are i.i.d vectors drawn from a product mean distribution, over a period of time the AI generates a large set of measurements xi called S that are grouped into two categories: the correct predictions M and the incorrect predictions Y. Once the network has made its predictions, a corrector can then be applied through centering S and Y by subtracting their means. The data is then regularized by applying the Kaiser rule to the resulting eigenmatrix and then whitened before being split into pairwise, positively correlated clusters. Each of these clusters produces a unique hyperplane and if any element x falls outside the region bounded by these lines then it is reported as an error. As a result of this methodology, a self-correcting recognition process is created that can identify fingerspelling from a variety of sign language and successfully identify the corresponding alphanumeric and what language the gesture originates from which no other neural network has been able to replicate.

Keywords: convolutional neural networks, deep learning, shallow correctors, sign language

Procedia PDF Downloads 74

Authors: Zachary Pan

Abstract:

Our general mission is to solve the problem of classifying images into different body part types and deciding if each of them is healthy or not. However, for now, we will determine healthiness for only one-sixth of the body parts, specifically the chest. We will detect pneumonia in X-ray scans of those chest images. With this type of AI, doctors can use it as a second opinion when they are taking CT or X-ray scans of their patients. Another ad-vantage of using this machine learning classifier is that it has no human weaknesses like fatigue. The overall ap-proach to this problem is to split the problem into two parts: first, classify the image, then determine if it is healthy. In order to classify the image into a specific body part class, the body parts dataset must be split into test and training sets. We can then use many models, like neural networks or logistic regression models, and fit them using the training set. Now, using the test set, we can obtain a realistic accuracy the models will have on images in the real world since these testing images have never been seen by the models before. In order to increase this testing accuracy, we can also apply many complex algorithms to the models, like multiplicative weight update. For the second part of the problem, to determine if the body part is healthy, we can have another dataset consisting of healthy and non-healthy images of the specific body part and once again split that into the test and training sets. We then use another neural network to train on those training set images and use the testing set to figure out its accuracy. We will do this process only for the chest images. A major conclusion reached is that convolutional neural networks are the most reliable and accurate at image classification. In classifying the images, the logistic regression model, the neural network, neural networks with multiplicative weight update, neural networks with the black box algorithm, and the convolutional neural network achieved 96.83 percent accuracy, 97.33 percent accuracy, 97.83 percent accuracy, 96.67 percent accuracy, and 98.83 percent accuracy, respectively. On the other hand, the overall accuracy of the model that de-termines if the images are healthy or not is around 78.37 percent accuracy.

Keywords: body part, healthcare, machine learning, neural networks

Procedia PDF Downloads 70

Authors: Mohamed Raouf Benmakrelouf, Joseph Rynkiewicz

Abstract:

Targeted maximum likelihood estimation (TMLE) is well-established method for causal effect estimation with desirable statistical properties. TMLE is a doubly robust maximum likelihood based approach that includes a secondary targeting step that optimizes the target statistical parameter. A causal interpretation of the statistical parameter requires assumptions of the Rubin causal framework. The causal effect of binary variable, E, on outcomes, Y, is defined in terms of comparisons between two potential outcomes as E[YE=1 − YE=0]. Our aim in this paper is to present an adaptation of TMLE methodology to estimate the causal effect of a non-binary categorical variable, providing a large application. We propose coding on the initial data in order to operate a binarization of the interest variable. For each category, we get a transformation of the non-binary interest variable into a binary variable, taking value 1 to indicate the presence of category (or group of categories) for an individual, 0 otherwise. Such a dummy variable makes it possible to have a pair of potential outcomes and oppose a category (or a group of categories) to another category (or a group of categories). Let E be a non-binary interest variable. We propose a complete disjunctive coding of our variable E. We transform the initial variable to obtain a set of binary vectors (dummy variables), E = (Ee : e ∈ {1, ..., |E|}), where each vector (variable), Ee, takes the value of 0 when its category is not present, and the value of 1 when its category is present, which allows to compute a pairwise-TMLE comparing difference in the outcome between one category and all remaining categories. In order to illustrate the application of our strategy, first, we present the implementation of TMLE to estimate the causal effect of non-binary variable on outcome using simulated data. Secondly, we apply our TMLE adaptation to survey data from the French Political Barometer (CEVIPOF), to estimate the causal effect of education level (A five-level variable) on a potential vote in favor of the French extreme right candidate Jean-Marie Le Pen. Counterfactual reasoning requires us to consider some causal questions (additional causal assumptions). Leading to different coding of E, as a set of binary vectors, E = (Ee : e ∈ {2, ..., |E|}), where each vector (variable), Ee, takes the value of 0 when the first category (reference category) is present, and the value of 1 when its category is present, which allows to apply a pairwise-TMLE comparing difference in the outcome between the first level (fixed) and each remaining level. We confirmed that the increase in the level of education decreases the voting rate for the extreme right party.

Keywords: statistical inference, causal inference, super learning, targeted maximum likelihood estimation

Procedia PDF Downloads 68

Authors: Abhisek Dash

Abstract:

This study had three neural networks, one for number segmentation, one for number detection and one for number recognition all of which are coupled to one another. All networks were trained on the MNIST dataset and were convolutional. It was assumed that the images had lighter background and darker foreground. The segmentation network took 28x28 images as input and had sixteen outputs. Segmentation training starts when a dark pixel is encountered. Taking a window(7x7) over that pixel as focus, the eight neighborhood of the focus was checked for further dark pixels. The segmentation network was then trained to move in those directions which had dark pixels. To this end the segmentation network had 16 outputs. They were arranged as “go east”, ”don’t go east ”, “go south east”, “don’t go south east”, “go south”, “don’t go south” and so on w.r.t focus window. The focus window was resized into a 28x28 image and the network was trained to consider those neighborhoods which had dark pixels. The neighborhoods which had dark pixels were pushed into a queue in a particular order. The neighborhoods were then popped one at a time stitched to the existing partial image of the number one at a time and trained on which neighborhoods to consider when the new partial image was presented. The above process was repeated until the image was fully covered by the 7x7 neighborhoods and there were no more uncovered black pixels. During testing the network scans and looks for the first dark pixel. From here on the network predicts which neighborhoods to consider and segments the image. After this step the group of neighborhoods are passed into the detection network. The detection network took 28x28 images as input and had two outputs denoting whether a number was detected or not. Since the ground truth of the bounds of a number was known during training the detection network outputted in favor of number not found until the bounds were not met and vice versa. The recognition network was a standard CNN that also took 28x28 images and had 10 outputs for recognition of numbers from 0 to 9. This network was activated only when the detection network votes in favor of number detected. The above methodology could segment connected and overlapping numbers. Additionally the recognition unit was only invoked when a number was detected which minimized false positives. It also eliminated the need for rules of thumb as segmentation is learned. The strategy can also be extended to other characters as well.

Keywords: convolutional neural networks, OCR, text detection, text segmentation

Procedia PDF Downloads 128

Authors: Hritwik Ghosh, Irfan Sadiq Rahat, Sachi Nandan Mohanty, J. V. R. Ravindra

Abstract:

In the rapidly evolving landscape of medical diagnostics, the early detection and accurate classification of skin cancer remain paramount for effective treatment outcomes. This research delves into the transformative potential of Artificial Intelligence (AI), specifically Deep Learning (DL), as a tool for discerning and categorizing various skin conditions. Utilizing a diverse dataset of 3,000 images representing nine distinct skin conditions, we confront the inherent challenge of class imbalance. This imbalance, where conditions like melanomas are over-represented, is addressed by incorporating class weights during the model training phase, ensuring an equitable representation of all conditions in the learning process. Our pioneering approach introduces a hybrid model, amalgamating the strengths of two renowned Convolutional Neural Networks (CNNs), VGG16 and ResNet50. These networks, pre-trained on the ImageNet dataset, are adept at extracting intricate features from images. By synergizing these models, our research aims to capture a holistic set of features, thereby bolstering classification performance. Preliminary findings underscore the hybrid model's superiority over individual models, showcasing its prowess in feature extraction and classification. Moreover, the research emphasizes the significance of rigorous data pre-processing, including image resizing, color normalization, and segmentation, in ensuring data quality and model reliability. In essence, this study illuminates the promising role of AI and DL in revolutionizing skin cancer diagnostics, offering insights into its potential applications in broader medical domains.

Keywords: artificial intelligence, machine learning, deep learning, skin cancer, dermatology, convolutional neural networks, image classification, computer vision, healthcare technology, cancer detection, medical imaging

Procedia PDF Downloads 42

Authors: Shin-Pin Tseng

Abstract:

Nowadays, numerous spectral amplitude coding (SAC) fiber-optic code-division-multiple-access (FO-CDMA) techniques were appealing due to their capable of providing moderate security and relieving the effects of multiuser interference (MUI). Nonetheless, the performance of the previous network is degraded due to fixed in-phase cross-correlation (IPCC) value. Based on the above problems, a new SAC FO-CDMA network using partial M-sequence (PMS) code is presented in this study. Because the proposed PMS code is originated from M-sequence code, the system using the PMS code could effectively suppress the effects of MUI. In addition, two-code keying (TCK) scheme can applied in the proposed SAC FO-CDMA network and enhance the whole network performance. According to the consideration of system flexibility, simple optical encoders/decoders (codecs) using fiber Bragg gratings (FBGs) were also developed. First, we constructed a diagram of the SAC FO-CDMA network, including (N/2-1) optical transmitters, (N/2-1) optical receivers, and one N×N star coupler for broadcasting transmitted optical signals to arrive at the input port of each optical receiver. Note that the parameter N for the PMS code was the code length. In addition, the proposed SAC network was using superluminescent diodes (SLDs) as light sources, which then can save a lot of system cost compared with the other FO-CDMA methods. For the design of each optical transmitter, it is composed of an SLD, one optical switch, and two optical encoders according to assigned PMS codewords. On the other hand, each optical receivers includes a 1 × 2 splitter, two optical decoders, and one balanced photodiode for mitigating the effect of MUI. In order to simplify the next analysis, the some assumptions were used. First, the unipolarized SLD has flat power spectral density (PSD). Second, the received optical power at the input port of each optical receiver is the same. Third, all photodiodes in the proposed network have the same electrical properties. Fourth, transmitting '1' and '0' has an equal probability. Subsequently, by taking the factors of phase‐induced intensity noise (PIIN) and thermal noise, the corresponding performance was displayed and compared with the performance of the previous SAC FO-CDMA networks. From the numerical result, it shows that the proposed network improved about 25% performance than that using other codes at BER=10-9. This is because the effect of PIIN was effectively mitigated and the received power was enhanced by two times. As a result, the SAC FO-CDMA network using PMS codes has an opportunity to apply in applications of the next-generation optical network.

Keywords: spectral amplitude coding, SAC, fiber-optic code-division multiple-access, FO-CDMA, partial M-sequence, PMS code, fiber Bragg grating, FBG

Procedia PDF Downloads 349

Authors: Po-Jung Huang, Chi-Ching Lee, Bertrand Chin-Ming Tan, Yuan-Ming Yeh, Julie Lichieh Chu, Tin-Wen Chen, Cheng-Yang Lee, Ruei-Chi Gan, Hsuan Liu, Petrus Tang

Abstract:

Whole-exome sequencing focuses on the protein coding regions of disease/cancer associated genes based on a priori knowledge is the most cost-effective method to study the association between genetic alterations and disease. Recent advances in high throughput sequencing technologies and proteomic techniques has provided an opportunity to integrate genomics and proteomics, allowing readily detectable mutated peptides corresponding to mutated genes. Since sequence database search is the most widely used method for protein identification using Mass spectrometry (MS)-based proteomics technology, a mutant proteome database is required to better approximate the real protein pool to improve disease-associated mutated protein identification. Large-scale whole exome/genome sequencing studies were launched by National Cancer Institute (NCI), Broad Institute, and The Cancer Genome Atlas (TCGA), which provide not only a comprehensive report on the analysis of coding variants in diverse samples cell lines but a invaluable resource for extensive research community. No existing database is available for the collection of mutant protein sequences related to the identified variants in these studies. CMPD is designed to address this issue, serving as a bridge between genomic data and proteomic studies and focusing on protein sequence-altering variations originated from both germline and cancer-associated somatic variations.

Keywords: TCGA, cancer, mutant, proteome

Procedia PDF Downloads 559

Authors: Ju-Hong Lee, Yi-Lin Shieh

Abstract:

Two-dimensional (2-D) quadrature mirror filter (QMF) banks have been widely considered for high-quality coding of image and video data at low bit rates. Without implementing subband coding, a 2-D QMF bank is required to have an exactly linear-phase response without magnitude distortion, i.e., the perfect reconstruction (PR) characteristics. The design problem of 2-D QMF banks with the PR characteristics has been considered in the literature for many years. This paper presents a transformation approach for designing 2-D two-channel QMF banks. Under a suitable one-dimensional (1-D) to two-dimensional (2-D) transformation with a specified decimation/interpolation matrix, the analysis and synthesis filters of the QMF bank are composed of 1-D causal and stable digital allpass filters (DAFs) and possess the 2-D doubly complementary half-band (DC-HB) property. This facilitates the design problem of the two-channel QMF banks by finding the real coefficients of the 1-D recursive DAFs. The design problem is formulated based on the minimax phase approximation for the 1-D DAFs. A novel objective function is then derived to obtain an optimization for 1-D minimax phase approximation. As a result, the problem of minimizing the objective function can be simply solved by using the well-known weighted least-squares (WLS) algorithm in the minimax (L∞) optimal sense. The novelty of the proposed design method is that the design procedure is very simple and the designed 2-D QMF bank achieves perfect magnitude response and possesses satisfactory phase response. Simulation results show that the proposed design method provides much better design performance and much less design complexity as compared with the existing techniques.

Keywords: Quincunx QMF bank, doubly complementary filter, digital allpass filter, WLS algorithm

Procedia PDF Downloads 204

Authors: Maxime Merheb, Rachel Matar

Abstract:

Retrieving ancient DNA sequences which in return permit the entire genome sequencing from fossils have extraordinarily improved in recent years, thanks to sequencing technology and other methodological advances. In any case, the quest to search for ancient DNA is still obstructed by the damage inflicted on DNA which accumulates after the death of a living organism. We can characterize this damage into three main categories: (i) Physical abnormalities such as strand breaks which lead to the presence of short DNA fragments. (ii) Modified bases (mainly cytosine deamination) which cause errors in the sequence due to an incorporation of a false nucleotide during DNA amplification. (iii) DNA modifications referred to as blocking lesions, will halt the PCR extension which in return will also affect the amplification and sequencing process. We can clearly see that the issues arising from breakage and coding errors were significantly decreased in recent years. Fast sequencing of short DNA fragments was empowered by platforms for high-throughput sequencing, most of the coding errors were uncovered to be the consequences of cytosine deamination which can be easily removed from the DNA using enzymatic treatment. The methodology to repair DNA sequences is still in development, it can be basically explained by the process of reintroducing cytosine rather than uracil. This technique is thus restricted to amplified DNA molecules. To eliminate any type of damage (particularly those that block PCR) is a process still pending the complete repair methodologies; DNA detection right after extraction is highly needed. Before using any resources into extensive, unreasonable and uncertain repair techniques, it is vital to distinguish between two possible hypotheses; (i) DNA is none existent to be amplified to begin with therefore completely un-repairable, (ii) the DNA is refractory to PCR and it is worth to be repaired and amplified. Hence, it is extremely important to develop a non-enzymatic technique to detect the most degraded DNA.

Keywords: ancient DNA, DNA barcodong, enzymatic repair, PCR

Procedia PDF Downloads 376

Authors: Ja-Keoung Koo, Kensuke Nakamura, Hyohun Kim, Dongwha Shin, Yeonseok Kim, Ji-Su Ahn, Byung-Woo Hong

Abstract:

The machine learning techniques based on a convolutional neural network (CNN) have been actively developed and successfully applied to a variety of image analysis tasks including reconstruction, noise reduction, resolution enhancement, segmentation, motion estimation, object recognition. The classical visual information processing that ranges from low level tasks to high level ones has been widely developed in the deep learning framework. It is generally considered as a challenging problem to derive visual interpretation from high dimensional imagery data. A CNN is a class of feed-forward artificial neural network that usually consists of deep layers the connections of which are established by a series of non-linear operations. The CNN architecture is known to be shift invariant due to its shared weights and translation invariance characteristics. However, it is often computationally intractable to optimize the network in particular with a large number of convolution layers due to a large number of unknowns to be optimized with respect to the training set that is generally required to be large enough to effectively generalize the model under consideration. It is also necessary to limit the size of convolution kernels due to the computational expense despite of the recent development of effective parallel processing machinery, which leads to the use of the constantly small size of the convolution kernels throughout the deep CNN architecture. However, it is often desired to consider different scales in the analysis of visual features at different layers in the network. Thus, we propose a CNN model where different sizes of the convolution kernels are applied at each layer based on the random projection. We apply random filters with varying sizes and associate the filter responses with scalar weights that correspond to the standard deviation of the random filters. We are allowed to use large number of random filters with the cost of one scalar unknown for each filter. The computational cost in the back-propagation procedure does not increase with the larger size of the filters even though the additional computational cost is required in the computation of convolution in the feed-forward procedure. The use of random kernels with varying sizes allows to effectively analyze image features at multiple scales leading to a better generalization. The robustness and effectiveness of the proposed CNN based on random kernels are demonstrated by numerical experiments where the quantitative comparison of the well-known CNN architectures and our models that simply replace the convolution kernels with the random filters is performed. The experimental results indicate that our model achieves better performance with less number of unknown weights. The proposed algorithm has a high potential in the application of a variety of visual tasks based on the CNN framework. Acknowledgement—This work was supported by the MISP (Ministry of Science and ICT), Korea, under the National Program for Excellence in SW (20170001000011001) supervised by IITP, and NRF-2014R1A2A1A11051941, NRF2017R1A2B4006023.

Keywords: deep learning, convolutional neural network, random kernel, random projection, dimensionality reduction, object recognition

Procedia PDF Downloads 258