Matching Items (20)
Filtering by
- All Subjects: Machine Learning
- All Subjects: Computer Engineering
- Creators: Berisha, Visar
- Member of: Theses and Dissertations
- Status: Published
Description
Our ability to understand networks is important to many applications, from the analysis and modeling of biological networks to analyzing social networks. Unveiling network dynamics allows us to make predictions and decisions. Moreover, network dynamics models have inspired new ideas for computational methods involving multi-agent cooperation, offering effective solutions for optimization tasks. This dissertation presents new theoretical results on network inference and multi-agent optimization, split into two parts -
The first part deals with modeling and identification of network dynamics. I study two types of network dynamics arising from social and gene networks. Based on the network dynamics, the proposed network identification method works like a `network RADAR', meaning that interaction strengths between agents are inferred by injecting `signal' into the network and observing the resultant reverberation. In social networks, this is accomplished by stubborn agents whose opinions do not change throughout a discussion. In gene networks, genes are suppressed to create desired perturbations. The steady-states under these perturbations are characterized. In contrast to the common assumption of full rank input, I take a laxer assumption where low-rank input is used, to better model the empirical network data. Importantly, a network is proven to be identifiable from low rank data of rank that grows proportional to the network's sparsity. The proposed method is applied to synthetic and empirical data, and is shown to offer superior performance compared to prior work. The second part is concerned with algorithms on networks. I develop three consensus-based algorithms for multi-agent optimization. The first method is a decentralized Frank-Wolfe (DeFW) algorithm. The main advantage of DeFW lies on its projection-free nature, where we can replace the costly projection step in traditional algorithms by a low-cost linear optimization step. I prove the convergence rates of DeFW for convex and non-convex problems. I also develop two consensus-based alternating optimization algorithms --- one for least square problems and one for non-convex problems. These algorithms exploit the problem structure for faster convergence and their efficacy is demonstrated by numerical simulations.
I conclude this dissertation by describing future research directions.
The first part deals with modeling and identification of network dynamics. I study two types of network dynamics arising from social and gene networks. Based on the network dynamics, the proposed network identification method works like a `network RADAR', meaning that interaction strengths between agents are inferred by injecting `signal' into the network and observing the resultant reverberation. In social networks, this is accomplished by stubborn agents whose opinions do not change throughout a discussion. In gene networks, genes are suppressed to create desired perturbations. The steady-states under these perturbations are characterized. In contrast to the common assumption of full rank input, I take a laxer assumption where low-rank input is used, to better model the empirical network data. Importantly, a network is proven to be identifiable from low rank data of rank that grows proportional to the network's sparsity. The proposed method is applied to synthetic and empirical data, and is shown to offer superior performance compared to prior work. The second part is concerned with algorithms on networks. I develop three consensus-based algorithms for multi-agent optimization. The first method is a decentralized Frank-Wolfe (DeFW) algorithm. The main advantage of DeFW lies on its projection-free nature, where we can replace the costly projection step in traditional algorithms by a low-cost linear optimization step. I prove the convergence rates of DeFW for convex and non-convex problems. I also develop two consensus-based alternating optimization algorithms --- one for least square problems and one for non-convex problems. These algorithms exploit the problem structure for faster convergence and their efficacy is demonstrated by numerical simulations.
I conclude this dissertation by describing future research directions.
ContributorsWai, Hoi To (Author) / Scaglione, Anna (Thesis advisor) / Berisha, Visar (Committee member) / Nedich, Angelia (Committee member) / Ying, Lei (Committee member) / Arizona State University (Publisher)
Created2017
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
Machine learning (ML) has played an important role in several modern technological innovations and has become an important tool for researchers in various fields of interest. Besides engineering, ML techniques have started to spread across various departments of study, like health-care, medicine, diagnostics, social science, finance, economics etc. These techniques require data to train the algorithms and model a complex system and make predictions based on that model. Due to development of sophisticated sensors it has become easier to collect large volumes of data which is used to make necessary hypotheses using ML. The promising results obtained using ML have opened up new opportunities of research across various departments and this dissertation is a manifestation of it. Here, some unique studies have been presented, from which valuable inference have been drawn for a real-world complex system. Each study has its own unique sets of motivation and relevance to the real world. An ensemble of signal processing (SP) and ML techniques have been explored in each study. This dissertation provides the detailed systematic approach and discusses the results achieved in each study. Valuable inferences drawn from each study play a vital role in areas of science and technology, and it is worth further investigation. This dissertation also provides a set of useful SP and ML tools for researchers in various fields of interest.
ContributorsDutta, Arindam (Author) / Bliss, Daniel W (Thesis advisor) / Berisha, Visar (Committee member) / Richmond, Christ (Committee member) / Corman, Steven (Committee member) / Arizona State University (Publisher)
Created2018
Description
Epilepsy affects numerous people around the world and is characterized by recurring seizures, prompting the ability to predict them so precautionary measures may be employed. One promising algorithm extracts spatiotemporal correlation based features from intracranial electroencephalography signals for use with support vector machines. The robustness of this methodology is tested through a sensitivity analysis. Doing so also provides insight about how to construct more effective feature vectors.
ContributorsMa, Owen (Author) / Bliss, Daniel (Thesis director) / Berisha, Visar (Committee member) / Barrett, The Honors College (Contributor) / Electrical Engineering Program (Contributor)
Created2015-05
Description
Many neurological disorders, especially those that result in dementia, impact speech and language production. A number of studies have shown that there exist subtle changes in linguistic complexity in these individuals that precede disease onset. However, these studies are conducted on controlled speech samples from a specific task. This thesis explores the possibility of using natural language processing in order to detect declining linguistic complexity from more natural discourse. We use existing data from public figures suspected (or at risk) of suffering from cognitive-linguistic decline, downloaded from the Internet, to detect changes in linguistic complexity. In particular, we focus on two case studies. The first case study analyzes President Ronald Reagan’s transcribed spontaneous speech samples during his presidency. President Reagan was diagnosed with Alzheimer’s disease in 1994, however my results showed declining linguistic complexity during the span of the 8 years he was in office. President George Herbert Walker Bush, who has no known diagnosis of Alzheimer’s disease, shows no decline in the same measures. In the second case study, we analyze transcribed spontaneous speech samples from the news conferences of 10 current NFL players and 18 non-player personnel since 2007. The non-player personnel have never played professional football. Longitudinal analysis of linguistic complexity showed contrasting patterns in the two groups. The majority (6 of 10) of current players showed decline in at least one measure of linguistic complexity over time. In contrast, the majority (11 out of 18) of non-player personnel showed an increase in at least one linguistic complexity measure.
ContributorsWang, Shuai (Author) / Berisha, Visar (Thesis advisor) / LaCross, Amy (Committee member) / Tong, Hanghang (Committee member) / Arizona State University (Publisher)
Created2016
Description
Information divergence functions, such as the Kullback-Leibler divergence or the Hellinger distance, play a critical role in statistical signal processing and information theory; however estimating them can be challenge. Most often, parametric assumptions are made about the two distributions to estimate the divergence of interest. In cases where no parametric model fits the data, non-parametric density estimation is used. In statistical signal processing applications, Gaussianity is usually assumed since closed-form expressions for common divergence measures have been derived for this family of distributions. Parametric assumptions are preferred when it is known that the data follows the model, however this is rarely the case in real-word scenarios. Non-parametric density estimators are characterized by a very large number of parameters that have to be tuned with costly cross-validation. In this dissertation we focus on a specific family of non-parametric estimators, called direct estimators, that bypass density estimation completely and directly estimate the quantity of interest from the data. We introduce a new divergence measure, the $D_p$-divergence, that can be estimated directly from samples without parametric assumptions on the distribution. We show that the $D_p$-divergence bounds the binary, cross-domain, and multi-class Bayes error rates and, in certain cases, provides provably tighter bounds than the Hellinger divergence. In addition, we also propose a new methodology that allows the experimenter to construct direct estimators for existing divergence measures or to construct new divergence measures with custom properties that are tailored to the application. To examine the practical efficacy of these new methods, we evaluate them in a statistical learning framework on a series of real-world data science problems involving speech-based monitoring of neuro-motor disorders.
ContributorsWisler, Alan (Author) / Berisha, Visar (Thesis advisor) / Spanias, Andreas (Thesis advisor) / Liss, Julie (Committee member) / Bliss, Daniel (Committee member) / Arizona State University (Publisher)
Created2017
Description
This work details the bootstrap estimation of a nonparametric information divergence measure, the Dp divergence measure, using a power law model. To address the challenge posed by computing accurate divergence estimates given finite size data, the bootstrap approach is used in conjunction with a power law curve to calculate an asymptotic value of the divergence estimator. Monte Carlo estimates of Dp are found for increasing values of sample size, and a power law fit is used to relate the divergence estimates as a function of sample size. The fit is also used to generate a confidence interval for the estimate to characterize the quality of the estimate. We compare the performance of this method with the other estimation methods. The calculated divergence is applied to the binary classification problem. Using the inherent relation between divergence measures and classification error rate, an analysis of the Bayes error rate of several data sets is conducted using the asymptotic divergence estimate.
ContributorsKadambi, Pradyumna Sanjay (Author) / Berisha, Visar (Thesis director) / Bliss, Daniel (Committee member) / Electrical Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
Description
Divergence functions are both highly useful and fundamental to many areas in information theory and machine learning, but require either parametric approaches or prior knowledge of labels on the full data set. This paper presents a method to estimate the divergence between two data sets in the absence of fully labeled data. This semi-labeled case is common in many domains where labeling data by hand is expensive or time-consuming, or wherever large data sets are present. The theory derived in this paper is demonstrated on a simulated example, and then applied to a feature selection and classification problem from pathological speech analysis.
ContributorsGilton, Davis Leland (Author) / Berisha, Visar (Thesis director) / Cochran, Douglas (Committee member) / Electrical Engineering Program (Contributor) / Barrett, The Honors College (Contributor)
Created2016-05
Description
This dissertation explores applications of machine learning methods in service of the design of screening tests, which are ubiquitous in applications from social work, to criminology, to healthcare. In the first part, a novel Bayesian decision theory framework is presented for designing tree-based adaptive tests. On an application to youth delinquency in Honduras, the method produces a 15-item instrument that is almost as accurate as a full-length 150+ item test. The framework includes specific considerations for the context in which the test will be administered, and provides uncertainty quantification around the trade-offs of shortening lengthy tests. In the second part, classification complexity is explored via theoretical and empirical results from statistical learning theory, information theory, and empirical data complexity measures. A simulation study that explicitly controls two key aspects of classification complexity is performed to relate the theoretical and empirical approaches. Throughout, a unified language and notation that formalizes classification complexity is developed; this same notation is used in subsequent chapters to discuss classification complexity in the context of a speech-based screening test. In the final part, the relative merits of task and feature engineering when designing a speech-based cognitive screening test are explored. Through an extensive classification analysis on a clinical speech dataset from patients with normal cognition and Alzheimer’s disease, the speech elicitation task is shown to have a large impact on test accuracy; carefully performed task and feature engineering are required for best results. A new framework for objectively quantifying speech elicitation tasks is introduced, and two methods are proposed for automatically extracting insights into the aspects of the speech elicitation task that are driving classification performance. The dissertation closes with recommendations for how to evaluate the obtained insights and use them to guide future design of speech-based screening tests.
ContributorsKrantsevich, Chelsea (Author) / Hahn, P. Richard (Thesis advisor) / Berisha, Visar (Committee member) / Lopes, Hedibert (Committee member) / Renaut, Rosemary (Committee member) / Zheng, Yi (Committee member) / Arizona State University (Publisher)
Created2023
Description
Much evidence has shown that first language (L1) plays an important role in the formation of L2 phonological system during second language (L2) learning process. This combines with the fact that different L1s have distinct phonological patterns to indicate the diverse L2 speech learning outcomes for speakers from different L1 backgrounds. This dissertation hypothesizes that phonological distances between accented speech and speakers' L1 speech are also correlated with perceived accentedness, and the correlations are negative for some phonological properties. Moreover, contrastive phonological distinctions between L1s and L2 will manifest themselves in the accented speech produced by speaker from these L1s. To test the hypotheses, this study comes up with a computational model to analyze the accented speech properties in both segmental (short-term speech measurements on short-segment or phoneme level) and suprasegmental (long-term speech measurements on word, long-segment, or sentence level) feature space. The benefit of using a computational model is that it enables quantitative analysis of L1's effect on accent in terms of different phonological properties. The core parts of this computational model are feature extraction schemes to extract pronunciation and prosody representation of accented speech based on existing techniques in speech processing field. Correlation analysis on both segmental and suprasegmental feature space is conducted to look into the relationship between acoustic measurements related to L1s and perceived accentedness across several L1s. Multiple regression analysis is employed to investigate how the L1's effect impacts the perception of foreign accent, and how accented speech produced by speakers from different L1s behaves distinctly on segmental and suprasegmental feature spaces. Results unveil the potential application of the methodology in this study to provide quantitative analysis of accented speech, and extend current studies in L2 speech learning theory to large scale. Practically, this study further shows that the computational model proposed in this study can benefit automatic accentedness evaluation system by adding features related to speakers' L1s.
ContributorsTu, Ming (Author) / Berisha, Visar (Thesis advisor) / Liss, Julie M (Committee member) / Zhou, Yi (Committee member) / Arizona State University (Publisher)
Created2018