Matching Items (19)
Filtering by
- All Subjects: deep learning
- Creators: Computer Science and Engineering Program
Description
Well-established model systems exist in four out of the seven major classes of vertebrates. These include the mouse, chicken, frog and zebrafish. Noticeably missing from this list is a reptilian model organism for comparative studies between the vertebrates and for studies of biological processes unique to reptiles. To help fill in this gap the green anole lizard, Anolis carolinensis, is being adapted as a model organism. Despite the recent release of the complete genomic sequence of the A. carolinensis, the lizard lacks some resources to aid researchers in their studies. Particularly, the lack of transcriptomic resources for lizard has made it difficult to identify genes complete with alternative splice forms and untranslated regions (UTRs). As part of this work the genome annotation for A. carolinensis was improved through next generation sequencing and assembly of the transcriptomes from 14 different adult and embryonic tissues. This revised annotation of the lizard will improve comparative studies between vertebrates, as well as studies within A. carolinensis itself, by providing more accurate gene models, which provide the bases for molecular studies. To demonstrate the utility of the improved annotations and reptilian model organism, the developmental process of somitogenesis in the lizard was analyzed and compared with other vertebrates. This study identified several key features both divergent and convergent between the vertebrates, which was not previously known before analysis of a reptilian model organism. The improved genome annotations have also allowed for molecular studies of tail regeneration in the lizard. With the annotation of 3' UTR sequences and next generation sequencing, it is now possible to do expressional studies of miRNA and predict their mRNA target transcripts at genomic scale. Through next generation small RNA sequencing and subsequent analysis, several differentially expressed miRNAs were identified in the regenerating tail, suggesting miRNA may play a key role in regulating this process in lizards. Through miRNA target prediction several key biological pathways were identified as potentially under the regulation of miRNAs during tail regeneration. In total, this work has both helped advance A. carolinensis as model system and displayed the utility of a reptilian model system.
ContributorsEckalbar, Walter L (Author) / Kusumi, Kenro (Thesis advisor) / Huentelman, Matthew (Committee member) / Rawls, Jeffery (Committee member) / Wilson-Rawls, Norma (Committee member) / Arizona State University (Publisher)
Created2012
Description
Induced pluripotent stem cells (iPSCs) are an intriguing approach for neurological disease modeling, because neural lineage-specific cell types that retain the donors' complex genetics can be established in vitro. The statistical power of these iPSC-based models, however, is dependent on accurate diagnoses of the somatic cell donors; unfortunately, many neurodegenerative diseases are commonly misdiagnosed in live human subjects. Postmortem histopathological examination of a donor's brain, combined with premortem clinical criteria, is often the most robust approach to correctly classify an individual as a disease-specific case or unaffected control. We describe the establishment of primary dermal fibroblasts cells lines from 28 autopsy donors. These fibroblasts were used to examine the proliferative effects of establishment protocol, tissue amount, biopsy site, and donor age. As proof-of-principle, iPSCs were generated from fibroblasts from a 75-year-old male, whole body donor, defined as an unaffected neurological control by both clinical and histopathological criteria. To our knowledge, this is the first study describing autopsy donor-derived somatic cells being used for iPSC generation and subsequent neural differentiation. This unique approach also enables us to compare iPSC-derived cell cultures to endogenous tissues from the same donor. We utilized RNA sequencing (RNA-Seq) to evaluate the transcriptional progression of in vitro-differentiated neural cells (over a timecourse of 0, 35, 70, 105 and 140 days), and compared this with donor-identical temporal lobe tissue. We observed in vitro progression towards the reference brain tissue, supported by (i) a significant increasing monotonic correlation between the days of our timecourse and the number of actively transcribed protein-coding genes and long intergenic non-coding RNAs (lincRNAs) (P < 0.05), consistent with the transcriptional complexity of the brain, (ii) an increase in CpG methylation after neural differentiation that resembled the epigenomic signature of the endogenous tissue, and (iii) a significant decreasing monotonic correlation between the days of our timecourse and the percent of in vitro to brain-tissue differences (P < 0.05) for tissue-specific protein-coding genes and all putative lincRNAs. These studies support the utility of autopsy donors' somatic cells for iPSC-based neurological disease models, and provide evidence that in vitro neural differentiation can result in physiologically progression.
ContributorsHjelm, Brooke E (Author) / Craig, David W. (Thesis advisor) / Wilson-Rawls, Norma J. (Thesis advisor) / Huentelman, Matthew J. (Committee member) / Mason, Hugh S. (Committee member) / Kusumi, Kenro (Committee member) / Arizona State University (Publisher)
Created2013
Description
This paper presents work that was done to create a system capable of facial expression recognition (FER) using deep convolutional neural networks (CNNs) and test multiple configurations and methods. CNNs are able to extract powerful information about an image using multiple layers of generic feature detectors. The extracted information can be used to understand the image better through recognizing different features present within the image. Deep CNNs, however, require training sets that can be larger than a million pictures in order to fine tune their feature detectors. For the case of facial expression datasets, none of these large datasets are available. Due to this limited availability of data required to train a new CNN, the idea of using naïve domain adaptation is explored. Instead of creating and using a new CNN trained specifically to extract features related to FER, a previously trained CNN originally trained for another computer vision task is used. Work for this research involved creating a system that can run a CNN, can extract feature vectors from the CNN, and can classify these extracted features. Once this system was built, different aspects of the system were tested and tuned. These aspects include the pre-trained CNN that was used, the layer from which features were extracted, normalization used on input images, and training data for the classifier. Once properly tuned, the created system returned results more accurate than previous attempts on facial expression recognition. Based on these positive results, naïve domain adaptation is shown to successfully leverage advantages of deep CNNs for facial expression recognition.
ContributorsEusebio, Jose Miguel Ang (Author) / Panchanathan, Sethuraman (Thesis director) / McDaniel, Troy (Committee member) / Venkateswara, Hemanth (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
Description
Emotion recognition in conversation has applications within numerous domains such as affective computing and medicine. Recent methods for emotion recognition jointly utilize conversational data over several modalities including audio, video, and text. However, state-of-the-art frameworks for this task do not focus on the feature extraction and feature fusion steps of this process. This thesis aims to improve the state-of-the-art method by incorporating two components to better accomplish these steps. By doing so, we are able to produce improved representations for the text modality and better model the relationships between all modalities. This paper proposes two methods which focus on these concepts and provide improved accuracy over the state-of-the-art framework for multimodal emotion recognition in dialogue.
ContributorsRawal, Siddharth (Author) / Baral, Chitta (Thesis director) / Shah, Shrikant (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
Accurate pose initialization and pose estimation are crucial requirements in on-orbit space assembly and various other autonomous on-orbit tasks. However, pose initialization and pose estimation are much more difficult to do accurately and consistently in space. This is primarily due to not only the variable lighting conditions present in space, but also the power requirements mandated by space-flyable hardware. This thesis investigates leveraging a deep learning approach for monocular one-shot pose initialization and pose estimation. A convolutional neural network was used to estimate the 6D pose of an assembly truss object. This network was trained by utilizing synthetic imagery generated from a simulation testbed. Furthermore, techniques to quantify model uncertainty of the deep learning model were investigated and applied in the task of in-space pose estimation and pose initialization. The feasibility of this approach on low-power computational platforms was also tested. The results demonstrate that accurate pose initialization and pose estimation can be conducted using a convolutional neural network. In addition, the results show that the model uncertainty can be obtained from the network. Lastly, the use of deep learning for pose initialization and pose estimation in addition with uncertainty quantification was demonstrated to be feasible on low-power compute platforms.
ContributorsKailas, Siva Maneparambil (Author) / Ben Amor, Heni (Thesis director) / Detry, Renaud (Committee member) / Economics Program in CLAS (Contributor) / School of Mathematical and Statistical Sciences (Contributor) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
Spotify, one of the most popular music streaming services, has many
algorithms for recommending new music to users. However, at the
core of their recommendations is the collaborative filtering algorithm,
which recommends music based on what other people with similar
tastes have listened to [1]. While this can produce highly relevant
content recommendations, it tends to promote only popular content
[2]. The popularity bias inherent in collaborative-filtering based
systems can overlook music that fits a user’s taste, simply because
nobody else is listening to it. One possible solution to this problem is
to recommend music based on features of the music itself, and
recommend songs which have similar features. Here, a method for
extracting high-level features representing the mood of a song is
presented, with the aim of tailoring music recommendations to an
individual's mood, and providing music recommendations with
diversity in popularity.
algorithms for recommending new music to users. However, at the
core of their recommendations is the collaborative filtering algorithm,
which recommends music based on what other people with similar
tastes have listened to [1]. While this can produce highly relevant
content recommendations, it tends to promote only popular content
[2]. The popularity bias inherent in collaborative-filtering based
systems can overlook music that fits a user’s taste, simply because
nobody else is listening to it. One possible solution to this problem is
to recommend music based on features of the music itself, and
recommend songs which have similar features. Here, a method for
extracting high-level features representing the mood of a song is
presented, with the aim of tailoring music recommendations to an
individual's mood, and providing music recommendations with
diversity in popularity.
ContributorsGomez, Luis Angel (Author) / Kevin, Burger (Thesis director) / Alberto, Hernández (Committee member) / Arts, Media and Engineering Sch T (Contributor) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
In recent years, the development of new Machine Learning models has allowed for new technological advancements to be introduced for practical use across the world. Multiple studies and experiments have been conducted to create new variations of Machine Learning models with different algorithms to determine if potential systems would prove to be successful. Even today, there are still many research initiatives that are continuing to develop new models in the hopes to discover potential solutions for problems such as autonomous driving or determining the emotional value from a single sentence. One of the current popular research topics for Machine Learning is the development of Facial Expression Recognition systems. These Machine Learning models focus on classifying images of human faces that are expressing different emotions through facial expressions. In order to develop effective models to perform Facial Expression Recognition, researchers have gone on to utilize Deep Learning models, which are a more advanced implementation of Machine Learning models, known as Neural Networks. More specifically, the use of Convolutional Neural Networks has proven to be the most effective models for achieving highly accurate results at classifying images of various facial expressions. Convolutional Neural Networks are Deep Learning models that are capable of processing visual data, such as images and videos, and can be used to identify various facial expressions. The purpose of this project, I focused on learning about the important concepts of Machine Learning, Deep Learning, and Convolutional Neural Networks to implement a Convolutional Neural Network that was previously developed by a recommended research paper.
ContributorsFrace, Douglas R (Author) / Demakethepalli Venkateswara, Hemanth Kumar (Thesis director) / McDaniel, Troy (Committee member) / Computer Science and Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2020-05
Description
Immunotherapy is an effective treatment for cancer which enables the patient's immune system to recognize tumor cells as pathogens. In order to design an individualized treatment, the t cell receptors (TCR) which bind to a tumor's unique antigens need to be determined. We created a convolutional neural network to predict the binding affinity between a given TCR and antigen to enable this.
ContributorsCai, Michael Ray (Author) / Lee, Heewook (Thesis director) / Meuth, Ryan (Committee member) / Computer Science and Engineering Program (Contributor, Contributor) / Barrett, The Honors College (Contributor)
Created2020-12
Description
Convolutional neural networks boast a myriad of applications in artificial intelligence, but one of the most common uses for such networks is image extraction. The ability of convolutional layers to extract and combine data features for the purpose of image analysis can be leveraged for pose estimation on an object - detecting the presence and attitude of corners and edges allows a convolutional neural network to identify how an object is positioned. This task can assist in working to grasp an object correctly in robotics applications, or to track an object more accurately in 3D space. However, the effectiveness of pose estimation may change based on properties of the object; the pose of a complex object, complexity being determined by internal occlusions, similar faces, etcetera, can be difficult to resolve.
This thesis is part of a collaboration between ASU’s Interactive Robotics Laboratory and NASA’s Jet Propulsion Laboratory. In this thesis, the training pipeline from Sharma’s paper “Pose Estimation for Non-Cooperative Spacecraft Rendezvous Using Convolutional Neural Networks” was modified to perform pose estimation on a complex object - specifically, a segment of a hollow truss. After initial attempts to replicate the architecture used in the paper and train solely on synthetic images, a combination of synthetic dataset generation and transfer learning on an ImageNet-pretrained AlexNet model was implemented to mitigate the difficulty of gathering large amounts of real-world data. Experimentation with pose estimation accuracy and hyperparameters of the model resulted in gradual test accuracy improvement, and future work is suggested to improve pose estimation for complex objects with some form of rotational symmetry.
This thesis is part of a collaboration between ASU’s Interactive Robotics Laboratory and NASA’s Jet Propulsion Laboratory. In this thesis, the training pipeline from Sharma’s paper “Pose Estimation for Non-Cooperative Spacecraft Rendezvous Using Convolutional Neural Networks” was modified to perform pose estimation on a complex object - specifically, a segment of a hollow truss. After initial attempts to replicate the architecture used in the paper and train solely on synthetic images, a combination of synthetic dataset generation and transfer learning on an ImageNet-pretrained AlexNet model was implemented to mitigate the difficulty of gathering large amounts of real-world data. Experimentation with pose estimation accuracy and hyperparameters of the model resulted in gradual test accuracy improvement, and future work is suggested to improve pose estimation for complex objects with some form of rotational symmetry.
ContributorsDsouza, Susanna Roshini (Author) / Ben Amor, Hani (Thesis director) / Maneparambil, Kailasnath (Committee member) / Computer Science and Engineering Program (Contributor, Contributor) / School of Mathematical and Statistical Sciences (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05
Description
Classical planning is a field of Artificial Intelligence concerned with allowing autonomous agents to make reasonable decisions in complex environments. This work investigates
the application of deep learning and planning techniques, with the aim of constructing generalized plans capable of solving multiple problem instances. We construct a Deep Neural Network that, given an abstract problem state, predicts both (i) the best action to be taken from that state and (ii) the generalized “role” of the object being manipulated. The neural network was tested on two classical planning domains: the blocks world domain and the logistic domain. Results indicate that neural networks are capable of making such
predictions with high accuracy, indicating a promising new framework for approaching generalized planning problems.
the application of deep learning and planning techniques, with the aim of constructing generalized plans capable of solving multiple problem instances. We construct a Deep Neural Network that, given an abstract problem state, predicts both (i) the best action to be taken from that state and (ii) the generalized “role” of the object being manipulated. The neural network was tested on two classical planning domains: the blocks world domain and the logistic domain. Results indicate that neural networks are capable of making such
predictions with high accuracy, indicating a promising new framework for approaching generalized planning problems.
ContributorsNakhleh, Julia Blair (Author) / Srivastava, Siddharth (Thesis director) / Fainekos, Georgios (Committee member) / Computer Science and Engineering Program (Contributor) / School of International Letters and Cultures (Contributor) / Barrett, The Honors College (Contributor)
Created2019-05