Matching Items (19)
Filtering by
- All Subjects: deep learning
- Creators: Seo, Jae-Sun
- Creators: Wu, Teresa
Description
Data imbalance and data noise often coexist in real world datasets. Data imbalance affects the learning classifier by degrading the recognition power of the classifier on the minority class, while data noise affects the learning classifier by providing inaccurate information and thus misleads the classifier. Because of these differences, data imbalance and data noise have been treated separately in the data mining field. Yet, such approach ignores the mutual effects and as a result may lead to new problems. A desirable solution is to tackle these two issues jointly. Noting the complementary nature of generative and discriminative models, this research proposes a unified model fusion based framework to handle the imbalanced classification with noisy dataset.
The phase I study focuses on the imbalanced classification problem. A generative classifier, Gaussian Mixture Model (GMM) is studied which can learn the distribution of the imbalance data to improve the discrimination power on imbalanced classes. By fusing this knowledge into cost SVM (cSVM), a CSG method is proposed. Experimental results show the effectiveness of CSG in dealing with imbalanced classification problems.
The phase II study expands the research scope to include the noisy dataset into the imbalanced classification problem. A model fusion based framework, K Nearest Gaussian (KNG) is proposed. KNG employs a generative modeling method, GMM, to model the training data as Gaussian mixtures and form adjustable confidence regions which are less sensitive to data imbalance and noise. Motivated by the K-nearest neighbor algorithm, the neighboring Gaussians are used to classify the testing instances. Experimental results show KNG method greatly outperforms traditional classification methods in dealing with imbalanced classification problems with noisy dataset.
The phase III study addresses the issues of feature selection and parameter tuning of KNG algorithm. To further improve the performance of KNG algorithm, a Particle Swarm Optimization based method (PSO-KNG) is proposed. PSO-KNG formulates model parameters and data features into the same particle vector and thus can search the best feature and parameter combination jointly. The experimental results show that PSO can greatly improve the performance of KNG with better accuracy and much lower computational cost.
The phase I study focuses on the imbalanced classification problem. A generative classifier, Gaussian Mixture Model (GMM) is studied which can learn the distribution of the imbalance data to improve the discrimination power on imbalanced classes. By fusing this knowledge into cost SVM (cSVM), a CSG method is proposed. Experimental results show the effectiveness of CSG in dealing with imbalanced classification problems.
The phase II study expands the research scope to include the noisy dataset into the imbalanced classification problem. A model fusion based framework, K Nearest Gaussian (KNG) is proposed. KNG employs a generative modeling method, GMM, to model the training data as Gaussian mixtures and form adjustable confidence regions which are less sensitive to data imbalance and noise. Motivated by the K-nearest neighbor algorithm, the neighboring Gaussians are used to classify the testing instances. Experimental results show KNG method greatly outperforms traditional classification methods in dealing with imbalanced classification problems with noisy dataset.
The phase III study addresses the issues of feature selection and parameter tuning of KNG algorithm. To further improve the performance of KNG algorithm, a Particle Swarm Optimization based method (PSO-KNG) is proposed. PSO-KNG formulates model parameters and data features into the same particle vector and thus can search the best feature and parameter combination jointly. The experimental results show that PSO can greatly improve the performance of KNG with better accuracy and much lower computational cost.
ContributorsHe, Miao (Author) / Wu, Teresa (Thesis advisor) / Li, Jing (Committee member) / Silva, Alvin (Committee member) / Borror, Connie (Committee member) / Arizona State University (Publisher)
Created2014
Description
In healthcare facilities, health information systems (HISs) are used to serve different purposes. The radiology department adopts multiple HISs in managing their operations and patient care. In general, the HISs that touch radiology fall into two categories: tracking HISs and archive HISs. Electronic Health Records (EHR) is a typical tracking HIS, which tracks the care each patient receives at multiple encounters and facilities. Archive HISs are typically specialized databases to store large-size data collected as part of the patient care. A typical example of an archive HIS is the Picture Archive and Communication System (PACS), which provides economical storage and convenient access to diagnostic images from multiple modalities. How to integrate such HISs and best utilize their data remains a challenging problem due to the disparity of HISs as well as high-dimensionality and heterogeneity of the data. My PhD dissertation research includes three inter-connected and integrated topics and focuses on designing integrated HISs and further developing statistical models and machine learning algorithms for process and patient care improvement.
Topic 1: Design of super-HIS and tracking of quality of care (QoC). My research developed an information technology that integrates multiple HISs in radiology, and proposed QoC metrics defined upon the data that measure various dimensions of care. The DDD assisted the clinical practices and enabled an effective intervention for reducing lengthy radiologist turnaround times for patients.
Topic 2: Monitoring and change detection of QoC data streams for process improvement. With the super-HIS in place, high-dimensional data streams of QoC metrics are generated. I developed a statistical model for monitoring high- dimensional data streams that integrated Singular Vector Decomposition (SVD) and process control. The algorithm was applied to QoC metrics data, and additionally extended to another application of monitoring traffic data in communication networks.
Topic 3: Deep transfer learning of archive HIS data for computer-aided diagnosis (CAD). The novelty of the CAD system is the development of a deep transfer learning algorithm that combines the ideas of transfer learning and multi- modality image integration under the deep learning framework. Our system achieved high accuracy in breast cancer diagnosis compared with conventional machine learning algorithms.
Topic 1: Design of super-HIS and tracking of quality of care (QoC). My research developed an information technology that integrates multiple HISs in radiology, and proposed QoC metrics defined upon the data that measure various dimensions of care. The DDD assisted the clinical practices and enabled an effective intervention for reducing lengthy radiologist turnaround times for patients.
Topic 2: Monitoring and change detection of QoC data streams for process improvement. With the super-HIS in place, high-dimensional data streams of QoC metrics are generated. I developed a statistical model for monitoring high- dimensional data streams that integrated Singular Vector Decomposition (SVD) and process control. The algorithm was applied to QoC metrics data, and additionally extended to another application of monitoring traffic data in communication networks.
Topic 3: Deep transfer learning of archive HIS data for computer-aided diagnosis (CAD). The novelty of the CAD system is the development of a deep transfer learning algorithm that combines the ideas of transfer learning and multi- modality image integration under the deep learning framework. Our system achieved high accuracy in breast cancer diagnosis compared with conventional machine learning algorithms.
ContributorsWang, Kun (Author) / Li, Jing (Thesis advisor) / Wu, Teresa (Committee member) / Pan, Rong (Committee member) / Zwart, Christine M. (Committee member) / Arizona State University (Publisher)
Created2018
Description
Deep neural networks (DNN) have shown tremendous success in various cognitive tasks, such as image classification, speech recognition, etc. However, their usage on resource-constrained edge devices has been limited due to high computation and large memory requirement.
To overcome these challenges, recent works have extensively investigated model compression techniques such as element-wise sparsity, structured sparsity and quantization. While most of these works have applied these compression techniques in isolation, there have been very few studies on application of quantization and structured sparsity together on a DNN model.
This thesis co-optimizes structured sparsity and quantization constraints on DNN models during training. Specifically, it obtains optimal setting of 2-bit weight and 2-bit activation coupled with 4X structured compression by performing combined exploration of quantization and structured compression settings. The optimal DNN model achieves 50X weight memory reduction compared to floating-point uncompressed DNN. This memory saving is significant since applying only structured sparsity constraints achieves 2X memory savings and only quantization constraints achieves 16X memory savings. The algorithm has been validated on both high and low capacity DNNs and on wide-sparse and deep-sparse DNN models. Experiments demonstrated that deep-sparse DNN outperforms shallow-dense DNN with varying level of memory savings depending on DNN precision and sparsity levels. This work further proposed a Pareto-optimal approach to systematically extract optimal DNN models from a huge set of sparse and dense DNN models. The resulting 11 optimal designs were further evaluated by considering overall DNN memory which includes activation memory and weight memory. It was found that there is only a small change in the memory footprint of the optimal designs corresponding to the low sparsity DNNs. However, activation memory cannot be ignored for high sparsity DNNs.
To overcome these challenges, recent works have extensively investigated model compression techniques such as element-wise sparsity, structured sparsity and quantization. While most of these works have applied these compression techniques in isolation, there have been very few studies on application of quantization and structured sparsity together on a DNN model.
This thesis co-optimizes structured sparsity and quantization constraints on DNN models during training. Specifically, it obtains optimal setting of 2-bit weight and 2-bit activation coupled with 4X structured compression by performing combined exploration of quantization and structured compression settings. The optimal DNN model achieves 50X weight memory reduction compared to floating-point uncompressed DNN. This memory saving is significant since applying only structured sparsity constraints achieves 2X memory savings and only quantization constraints achieves 16X memory savings. The algorithm has been validated on both high and low capacity DNNs and on wide-sparse and deep-sparse DNN models. Experiments demonstrated that deep-sparse DNN outperforms shallow-dense DNN with varying level of memory savings depending on DNN precision and sparsity levels. This work further proposed a Pareto-optimal approach to systematically extract optimal DNN models from a huge set of sparse and dense DNN models. The resulting 11 optimal designs were further evaluated by considering overall DNN memory which includes activation memory and weight memory. It was found that there is only a small change in the memory footprint of the optimal designs corresponding to the low sparsity DNNs. However, activation memory cannot be ignored for high sparsity DNNs.
ContributorsSrivastava, Gaurav (Author) / Seo, Jae-Sun (Thesis advisor) / Chakrabarti, Chaitali (Committee member) / Berisha, Visar (Committee member) / Arizona State University (Publisher)
Created2018
Description
Deep learning (DL) has proved itself be one of the most important developements till date with far reaching impacts in numerous fields like robotics, computer vision, surveillance, speech processing, machine translation, finance, etc. They are now widely used for countless applications because of their ability to generalize real world data, robustness to noise in previously unseen data and high inference accuracy. With the ability to learn useful features from raw sensor data, deep learning algorithms have out-performed tradinal AI algorithms and pushed the boundaries of what can be achieved with AI. In this work, we demonstrate the power of deep learning by developing a neural network to automatically detect cough instances from audio recorded in un-constrained environments. For this, 24 hours long recordings from 9 dierent patients is collected and carefully labeled by medical personel. A pre-processing algorithm is proposed to convert event based cough dataset to a more informative dataset with start and end of coughs and also introduce data augmentation for regularizing the training procedure. The proposed neural network achieves 92.3% leave-one-out accuracy on data captured in real world.
Deep neural networks are composed of multiple layers that are compute/memory intensive. This makes it difficult to execute these algorithms real-time with low power consumption using existing general purpose computers. In this work, we propose hardware accelerators for a traditional AI algorithm based on random forest trees and two representative deep convolutional neural networks (AlexNet and VGG). With the proposed acceleration techniques, ~ 30x performance improvement was achieved compared to CPU for random forest trees. For deep CNNS, we demonstrate that much higher performance can be achieved with architecture space exploration using any optimization algorithms with system level performance and area models for hardware primitives as inputs and goal of minimizing latency with given resource constraints. With this method, ~30GOPs performance was achieved for Stratix V FPGA boards.
Hardware acceleration of DL algorithms alone is not always the most ecient way and sucient to achieve desired performance. There is a huge headroom available for performance improvement provided the algorithms are designed keeping in mind the hardware limitations and bottlenecks. This work achieves hardware-software co-optimization for Non-Maximal Suppression (NMS) algorithm. Using the proposed algorithmic changes and hardware architecture
With CMOS scaling coming to an end and increasing memory bandwidth bottlenecks, CMOS based system might not scale enough to accommodate requirements of more complicated and deeper neural networks in future. In this work, we explore RRAM crossbars and arrays as compact, high performing and energy efficient alternative to CMOS accelerators for deep learning training and inference. We propose and implement RRAM periphery read and write circuits and achieved ~3000x performance improvement in online dictionary learning compared to CPU.
This work also examines the realistic RRAM devices and their non-idealities. We do an in-depth study of the effects of RRAM non-idealities on inference accuracy when a pretrained model is mapped to RRAM based accelerators. To mitigate this issue, we propose Random Sparse Adaptation (RSA), a novel scheme aimed at tuning the model to take care of the faults of the RRAM array on which it is mapped. Our proposed method can achieve inference accuracy much higher than what traditional Read-Verify-Write (R-V-W) method could achieve. RSA can also recover lost inference accuracy 100x ~ 1000x faster compared to R-V-W. Using 32-bit high precision RSA cells, we achieved ~10% higher accuracy using fautly RRAM arrays compared to what can be achieved by mapping a deep network to an 32 level RRAM array with no variations.
Deep neural networks are composed of multiple layers that are compute/memory intensive. This makes it difficult to execute these algorithms real-time with low power consumption using existing general purpose computers. In this work, we propose hardware accelerators for a traditional AI algorithm based on random forest trees and two representative deep convolutional neural networks (AlexNet and VGG). With the proposed acceleration techniques, ~ 30x performance improvement was achieved compared to CPU for random forest trees. For deep CNNS, we demonstrate that much higher performance can be achieved with architecture space exploration using any optimization algorithms with system level performance and area models for hardware primitives as inputs and goal of minimizing latency with given resource constraints. With this method, ~30GOPs performance was achieved for Stratix V FPGA boards.
Hardware acceleration of DL algorithms alone is not always the most ecient way and sucient to achieve desired performance. There is a huge headroom available for performance improvement provided the algorithms are designed keeping in mind the hardware limitations and bottlenecks. This work achieves hardware-software co-optimization for Non-Maximal Suppression (NMS) algorithm. Using the proposed algorithmic changes and hardware architecture
With CMOS scaling coming to an end and increasing memory bandwidth bottlenecks, CMOS based system might not scale enough to accommodate requirements of more complicated and deeper neural networks in future. In this work, we explore RRAM crossbars and arrays as compact, high performing and energy efficient alternative to CMOS accelerators for deep learning training and inference. We propose and implement RRAM periphery read and write circuits and achieved ~3000x performance improvement in online dictionary learning compared to CPU.
This work also examines the realistic RRAM devices and their non-idealities. We do an in-depth study of the effects of RRAM non-idealities on inference accuracy when a pretrained model is mapped to RRAM based accelerators. To mitigate this issue, we propose Random Sparse Adaptation (RSA), a novel scheme aimed at tuning the model to take care of the faults of the RRAM array on which it is mapped. Our proposed method can achieve inference accuracy much higher than what traditional Read-Verify-Write (R-V-W) method could achieve. RSA can also recover lost inference accuracy 100x ~ 1000x faster compared to R-V-W. Using 32-bit high precision RSA cells, we achieved ~10% higher accuracy using fautly RRAM arrays compared to what can be achieved by mapping a deep network to an 32 level RRAM array with no variations.
ContributorsMohanty, Abinash (Author) / Cao, Yu (Thesis advisor) / Seo, Jae-Sun (Committee member) / Vrudhula, Sarma (Committee member) / Chakrabarti, Chaitali (Committee member) / Arizona State University (Publisher)
Created2018
Description
The current Enterprise Requirements and Acquisition Model (ERAM), a discrete event simulation of the major tasks and decisions within the DoD acquisition system, identifies several what-if intervention strategies to improve program completion time. However, processes that contribute to the program acquisition completion time were not explicitly identified in the simulation study. This research seeks to determine the acquisition processes that contribute significantly to total simulated program time in the acquisition system for all programs reaching Milestone C. Specifically, this research examines the effect of increased scope management, technology maturity, and decreased variation and mean process times in post-Design Readiness Review contractor activities by performing additional simulation analyses. Potential policies are formulated from the results to further improve program acquisition completion time.
ContributorsWorger, Danielle Marie (Author) / Wu, Teresa (Thesis director) / Shunk, Dan (Committee member) / Wirthlin, J. Robert (Committee member) / Industrial, Systems (Contributor) / Barrett, The Honors College (Contributor)
Created2013-05
Description
In recent years, Artificial Intelligence (AI) (e.g., Deep Neural Networks (DNNs), Transformer) has shown great success in real-world applications due to its superior performance in various cognitive tasks. The impressive performance achieved by AI models normally accompanies the cost of enormous model size and high computational complexity, which significantly hampers their implementation on resource-limited Cyber-Physical Systems (CPS), Internet-of-Things (IoT), or Edge systems due to their tightly constrained energy, computing, size, and memory budget. Thus, the urgent demand for enhancing the \textbf{Efficiency} of DNN has drawn significant research interests across various communities.
Motivated by the aforementioned concerns, this doctoral research has been mainly focusing on Enabling Deep Learning at Edge: From Efficient and Dynamic Inference to On-Device Learning. Specifically, from the inference perspective, this dissertation begins by investigating a hardware-friendly model compression method that effectively reduces the size of AI model while simultaneously achieving improved speed on edge devices. Additionally, due to the fact that diverse resource constraints of different edge devices, this dissertation further explores dynamic inference, which allows for real-time tuning of inference model size, computation, and latency to accommodate the limitations of each edge device. Regarding efficient on-device learning, this dissertation starts by analyzing memory usage during transfer learning training. Based on this analysis, a novel framework called "Reprogramming Network'' (Rep-Net) is introduced that offers a fresh perspective on the on-device transfer learning problem. The Rep-Net enables on-device transferlearning by directly learning to reprogram the intermediate features of a pre-trained model. Lastly, this dissertation studies an efficient continual learning algorithm that facilitates learning multiple tasks without the risk of forgetting previously acquired knowledge. In practice, through the exploration of task correlation, an interesting phenomenon is observed that the intermediate features are highly correlated between tasks with the self-supervised pre-trained model. Building upon this observation, a novel approach called progressive task-correlated layer freezing is proposed to gradually freeze a subset of layers with the highest correlation ratios for each task leading to training efficiency.
ContributorsYang, Li (Author) / Fan, Deliang (Thesis advisor) / Seo, Jae-Sun (Committee member) / Zhang, Junshan (Committee member) / Cao, Yu (Committee member) / Arizona State University (Publisher)
Created2023
Description
Understanding customer preference is crucial for new product planning and marketing decisions. This thesis explores how historical data can be leveraged to understand and predict customer preference. This thesis presents a decision support framework that provides a holistic view on customer preference by following a two-phase procedure. Phase-1 uses cluster analysis to create product profiles based on which customer profiles are derived. Phase-2 then delves deep into each of the customer profiles and investigates causality behind their preference using Bayesian networks. This thesis illustrates the working of the framework using the case of Intel Corporation, world’s largest semiconductor manufacturing company.
ContributorsRam, Sudarshan Venkat (Author) / Kempf, Karl G. (Thesis advisor) / Wu, Teresa (Thesis advisor) / Ju, Feng (Committee member) / Arizona State University (Publisher)
Created2017
Description
Nonalcoholic Steatohepatitis (NASH) is a severe form of Nonalcoholic fatty liverdisease, that is caused due to excessive calorie intake, sedentary lifestyle and in the
absence of severe alcohol consumption. It is widely prevalent in the United States
and in many other developed countries, affecting up to 25 percent of the population.
Due to being asymptotic, it usually goes unnoticed and may lead to liver failure if
not treated at the right time.
Currently, liver biopsy is the gold standard to diagnose NASH, but being an
invasive procedure, it comes with it's own complications along with the inconvenience
of sampling repeated measurements over a period of time. Hence, noninvasive
procedures to assess NASH are urgently required. Magnetic Resonance Elastography
(MRE) based Shear Stiffness and Loss Modulus along with Magnetic Resonance
Imaging based proton density fat fraction have been successfully combined to predict
NASH stages However, their role in the prediction of disease progression still remains
to be investigated.
This thesis thus looks into combining features from serial MRE observations to
develop statistical models to predict NASH progression. It utilizes data from an experiment
conducted on male mice to develop progressive and regressive NASH and
trains ordinal models, ordered probit regression and ordinal forest on labels generated
from a logistic regression model. The models are assessed on histological data collected
at the end point of the experiment. The models developed provide a framework
to utilize a non-invasive tool to predict NASH disease progression.
ContributorsDeshpande, Eeshan (Author) / Ju, Feng (Thesis advisor) / Wu, Teresa (Committee member) / Yan, Hao (Committee member) / Arizona State University (Publisher)
Created2021
Description
In this thesis, the applications of deep learning in the analysis, detection and classification of medical imaging datasets were studied, with a focus on datasets having a limited sample size. A combined machine learning-deep learning model was designed to classify one small dataset, prostate cancer provided by Mayo Clinic, Arizona. Deep learning model was implemented to extract imaging features followed by machine learning classifier for prostate cancer diagnosis. The results were compared against models trained on texture-based features, namely gray level co-occurrence matrix (GLCM) and Gabor. Some of the challenges of performing diagnosis on medical imaging datasets with limited sample sizes, have been identified. Lastly, a set of future works have been proposed. Keywords: Deep learning, radiology, transfer learning, convolutional neural network.
ContributorsSarkar, Suryadipto (Author) / Wu, Teresa (Thesis advisor) / Papandreou-Suppappola, Antonia (Committee member) / Silva, Alvin (Committee member) / Arizona State University (Publisher)
Created2021
Description
Over the past few decades, medical imaging is becoming important in medicine for disease diagnosis, prognosis, treatment assessment and health monitoring. As medical imaging has progressed, imaging biomarkers are being rapidly developed for early diagnosis and staging of disease. Detecting and segmenting objects from images are often the first steps in quantitative measurement of these biomarkers. While large objects can often be automatically or semi-automatically delineated, segmenting small objects (blobs) is challenging. The small object of particular interest in this dissertation are glomeruli from kidney magnetic resonance (MR) images. This problem has its unique challenges. First of all, the size of glomeruli is extremely small and very similar with noises from images. Second, there are massive of glomeruli in kidney, e.g. over 1 million glomeruli in human kidney, and the intensity distribution is heterogenous. A third recognized issue is that a large portion of glomeruli are overlapping and touched in images. The goal of this dissertation is to develop computational algorithms to identify and discover glomeruli related imaging biomarkers. The first phase is to develop a U-net joint with Hessian based Difference of Gaussians (UH-DoG) blob detector. Joining effort from deep learning alleviates the over-detection issue from Hessian analysis. Next, as extension of UH-DoG, a small blob detector using Bi-Threshold Constrained Adaptive Scales (BTCAS) is proposed. Deep learning is treated as prior of Difference of Gaussian (DoG) to improve its efficiency. By adopting BTCAS, under-segmentation issue of deep learning is addressed. The second phase is to develop a denoising convexity-consistent Blob Generative Adversarial Network (BlobGAN). BlobGAN could achieve high denoising performance and selectively denoise the image without affecting the blobs. These detectors are validated on datasets of 2D fluorescent images, 3D synthetic images, 3D MR (18 mice, 3 humans) images and proved to be outperforming the competing detectors. In the last phase, a Fréchet Descriptors Distance based Coreset approach (FDD-Coreset) is proposed for accelerating BlobGAN’s training. Experiments have shown that BlobGAN trained on FDD-Coreset not only significantly reduces the training time, but also achieves higher denoising performance and maintains approximate performance of blob identification compared with training on entire dataset.
ContributorsXu, Yanzhe (Author) / Wu, Teresa (Thesis advisor) / Iquebal, Ashif (Committee member) / Yan, Hao (Committee member) / Beeman, Scott (Committee member) / Arizona State University (Publisher)
Created2022