You are here
Search results
(1 - 20 of 154)
Pages
- Title
- Robust multi-task learning algorithms for predictive modeling of spatial and temporal data
- Creator
- Liu, Xi (Graduate of Michigan State University)
- Date
- 2019
- Collection
- Electronic Theses & Dissertations
- Description
-
"Recent years have witnessed the significant growth of spatial and temporal data generated from various disciplines, including geophysical sciences, neuroscience, economics, criminology, and epidemiology. Such data have been extensively used to train spatial and temporal models that can make predictions either at multiple locations simultaneously or along multiple forecasting horizons (lead times). However, training an accurate prediction model in these domains can be challenging especially...
Show more"Recent years have witnessed the significant growth of spatial and temporal data generated from various disciplines, including geophysical sciences, neuroscience, economics, criminology, and epidemiology. Such data have been extensively used to train spatial and temporal models that can make predictions either at multiple locations simultaneously or along multiple forecasting horizons (lead times). However, training an accurate prediction model in these domains can be challenging especially when there are significant noise and missing values or limited training examples available. The goal of this thesis is to develop novel multi-task learning frameworks that can exploit the spatial and/or temporal dependencies of the data to ensure robust predictions in spite of the data quality and scarcity problems. The first framework developed in this dissertation is designed for multi-task classification of time series data. Specifically, the prediction task here is to continuously classify activities of a human subject based on the multi-modal sensor data collected in a smart home environment. As the classes exhibit strong spatial and temporal dependencies, this makes it an ideal setting for applying a multi-task learning approach. Nevertheless, since the type of sensors deployed often vary from one room (location) to another, this introduces a structured missing value problem, in which blocks of sensor data could be missing when a subject moves from one room to another. To address this challenge, a probabilistic multi-task classification framework is developed to jointly model the activity recognition tasks from all the rooms, taking into account the block-missing value problem. The framework also learns the transitional dependencies between classes to improve its overall prediction accuracy. The second framework is developed for the multi-location time series forecasting problem. Although multi-task learning has been successfully applied to many time series forecasting applications such as climate prediction, conventional approaches aim to minimize only the point-wise residual error of their predictions instead of considering how well their models fit the overall distribution of the response variable. As a result, their predicted distribution may not fully capture the true distribution of the data. In this thesis, a novel distribution-preserving multi-task learning framework is proposed for the multi-location time series forecasting problem. The framework uses a non-parametric density estimation approach to fit the distribution of the response variable and employs an L2-distance function to minimize the divergence between the predicted and true distributions. The third framework proposed in this dissertation is for the multi-step-ahead (long-range) time series prediction problem with application to ensemble forecasting of sea surface temperature. Specifically, our goal is to effectively combine the forecasts generated by various numerical models at different lead times to obtain more precise predictions. Towards this end, a multi-task deep learning framework based on a hierarchical LSTM architecture is proposed to jointly model the ensemble forecasts of different models, taking into account the temporal dependencies between forecasts at different lead times. Experiments performed on 29-year sea surface temperature data from North American Multi-Model Ensemble (NAMME) demonstrate that the proposed architecture significantly outperforms standard LSTM and other MTL approaches."--Pages ii-iii.
Show less
- Title
- Improving spectrum efficiency in heterogeneous wireless networks
- Creator
- Liu, Chin-Jung
- Date
- 2018
- Collection
- Electronic Theses & Dissertations
- Description
-
Over the past decades, the bandwidth-intensive applications that are previously confined to wired networks are now migrating to wireless networks. This trend has brought unprecedented high demand for wireless bandwidth. The wireless traffic is destined to dominate the Internet traffic in the future, but many of the popular wireless spectrum bands, especially the cellular and ISM bands, are already congested. On the other hand, some other wireless technologies, such as TV bands, often do not...
Show moreOver the past decades, the bandwidth-intensive applications that are previously confined to wired networks are now migrating to wireless networks. This trend has brought unprecedented high demand for wireless bandwidth. The wireless traffic is destined to dominate the Internet traffic in the future, but many of the popular wireless spectrum bands, especially the cellular and ISM bands, are already congested. On the other hand, some other wireless technologies, such as TV bands, often do not fully utilize their spectrum. However, the spectrum allocation is tightly regulated by the authority and adjusting the allocation is extremely difficult. The uneven utilization and the rigid regulation have led to the proposal of heterogeneous wireless networks, including cognitive radio networks (CRN) and heterogeneous cellular networks (HetNet). The CRNs that usually operate on different technologies from the spectrum owner attempt to reuse the idle spectrum (i.e., white space) from the owner, while HetNets attempt to improve spectrum utilization by smallcells. This dissertation addresses some of the challenging problems in these heterogeneous wireless networks.In CRNs, the secondary users (SU) are allowed to access the white spaces opportunistically as long as the SUs do not interfere with the primary users (PU, i.e., the spectrum owner). The CRN provides a promising means to improve spectral efficiency, which also introduces a set of new research challenges. We identify and discuss two problems in CRNs, namely non-contiguous control channel establishment and k-protected routing protocol design. The first problem deals with the need from SUs for a channel to transfer control information. Most existing approaches are channel-hopping (CH) based, which is inapplicable to NC-OFDM. We propose an efficient method for guaranteed NC-OFDM-based control channel establishment by utilizing short pulses on OFDM subcarriers. The results show that the time needed for establishing control channel is lower than that of CH-based approaches. The second problem deals with the interruption to a routing path in a CRN when a PU becomes active again. Existing reactive approaches that try to seek for an alternative route after PU returns suffer from potential long delay and possible interruption if an alternative cannot be found. We propose a k-protected routing protocol that builds routing paths with preassigned backups that are guaranteed to sustain from k returning PUs without being interrupted. Our result shows that the k-protected routing paths are never interrupted even when k PUs return, and have significantly shorter backup activation delays.HetNets formed by smallcells with different sizes of coverage and macrocells have been proposed to satisfy increased bandwidth demand with the limited and crowded wireless spectrum. Since the smallcells and macrocells operate on the same frequency, interference becomes a critical issue. Detecting and mitigating interference are two of the challenges introduced by HetNets. We first study the interference identification problem. Existing interference identification approaches often regard more cells as interferers than necessary. We propose to identify interference by analyzing the received patterns observed by the mobile stations. The result shows that our approach identifies all true interferers and excludes most non-interfering cells. The second research problem in HetNets is to provide effective solutions to mitigate the interference. The interference mitigation approaches in the literature mainly try to avoid interference, such as resource isolation that leads to significantly fewer resources, or power control that sacrifices signal quality and coverage. Instead of conservatively avoiding interference, we propose to mitigate the interference by precanceling the interfering signals from known interferers. With precancellation, the same set of resources can be shared between cells and thus throughput is improved.This dissertation addresses several challenges in heterogeneous wireless networks, including CRNs and HetNets. The proposed non-contiguous control channel protocol and k-protected routing protocol for CRNs can significantly improve the feasibility of CRNs in future wireless network applications. The proposed interference identification and interference precancellation approaches can effectively mitigate the interference and improve the throughput and spectrum utilization in HetNets. This dissertation aims at breaking the barriers for supporting heterogeneous wireless networks to improve the utilization of the precious and limited wireless spectrum.
Show less
- Title
- A framework for combining ancillary information with primary biometric traits
- Creator
- Ding, Yaohui
- Date
- 2018
- Collection
- Electronic Theses & Dissertations
- Description
-
"Biometric systems recognize individuals based on their biological attributes such as faces, fingerprints and iris. However, in several scenarios, additional ancillary information such as the biographic and demographic information of a user (e.g., name, gender, age, ethnicity), or the image quality of the biometric sample, anti-spoofing measurements, etc. may be available. While previous literature has studied the impact of such ancillary information on biometric system performance, there is...
Show more"Biometric systems recognize individuals based on their biological attributes such as faces, fingerprints and iris. However, in several scenarios, additional ancillary information such as the biographic and demographic information of a user (e.g., name, gender, age, ethnicity), or the image quality of the biometric sample, anti-spoofing measurements, etc. may be available. While previous literature has studied the impact of such ancillary information on biometric system performance, there is limited work on systematically incorporating them into the biometric matching framework. In this dissertation, we develop a principled framework to combine ancillary information with biometric match scores. The incorporation of ancillary information raises several challenges. Firstly, ancillary information such as gender, ethnicity and other demographic attributes lack distinctiveness and can be used to distinguish population groups rather than individuals. Secondly, ancillary information such as image quality and anti-spoof measurements may have different numerical ranges and interpretations. Further, most of the ancillary information cannot be automatically extracted without errors. Even the direct collection of ancillary information from subjects may be susceptible to transcription errors (e.g., errors in entering the data). Thirdly, the relationships between ancillary attributes and biometric traits may not be evident. In this regard, this dissertation makes three contributions. The first contribution entails the design of a Bayesian Belief Network (BBN) to model the relationship between biometric scores and ancillary factors, and exploiting the ensuing structure in a fusion framework. The ancillary information considered by the network includes image quality and anti-spoof measures. Experiments convey the importance of explicitly incorporating such information in a biometric system. The second contribution is the design of a Generalized Additive Model (GAM) that uses spline functions to model the correlation between match scores and ancillary attributes, and then learns a transformation function to normalize the match scores prior to fusion. The resulting framework can also be used to predict in advance if fusing match scores with certain demographic attributes is beneficial in the context of a specific biometric matcher. Experiments indicate that the proposed method can be used to significantly improve the recognition accuracy of state-of-the-art face matchers. The third contribution is the design of an ensemble of One Class Support Vector Machines (OC-SVMs) to combine multiple anti-spoofing measurements in order to mitigate the concerns associated with the issue of "imbalanced training sets" and "insufficient spoof samples" encountered by conventional anti-spoofing algorithms. In the proposed method, the spoof detection problem is formulated as a one-class problem, where the focus is on modeling a real fingerprint using multiple feature sets. The one-class classifiers corresponding to these multiple feature sets are then combined to generate a single classifier for spoof detection. Experimental results convey the importance of this technique in detecting spoofs made of materials that were not included in the training data. In summary, this dissertation seeks to advance our understanding of systematically exploiting ancillary information in designing effective biometric recognition systems by developing and evaluating multiple statistical models."--Pages ii-iii.
Show less
- Title
- Modeling physical causality of action verbs for grounded language understanding
- Creator
- Gao, Qiaozi
- Date
- 2019
- Collection
- Electronic Theses & Dissertations
- Description
-
Building systems that can understand and communicate through human natural language is one of the ultimate goals in AI. Decades of natural language processing research has been mainly focused on learning from large amounts of language corpora. However, human communication relies on a significant amount of unverbalized information, which is often referred as commonsense knowledge. This type of knowledge allows us to understand each other's intention, to connect language with concepts in the...
Show moreBuilding systems that can understand and communicate through human natural language is one of the ultimate goals in AI. Decades of natural language processing research has been mainly focused on learning from large amounts of language corpora. However, human communication relies on a significant amount of unverbalized information, which is often referred as commonsense knowledge. This type of knowledge allows us to understand each other's intention, to connect language with concepts in the world, and to make inference based on what we hear or read. Commonsense knowledge is generally shared among cognitive capable individuals, thus it is rarely stated in human language. This makes it very difficult for artificial agents to acquire commonsense knowledge from language corpora. To address this problem, this dissertation investigates the acquisition of commonsense knowledge, especially knowledge related to basic actions upon the physical world and how that influences language processing and grounding.Linguistics studies have shown that action verbs often denote some change of state (CoS) as the result of an action. For example, the result of "slice a pizza" is that the state of the object (pizza) changes from one big piece to several smaller pieces. However, the causality of action verbs and its potential connection with the physical world has not been systematically explored. Artificial agents often do not have this kind of basic commonsense causality knowledge, which makes it difficult for these agents to work with humans and to reason, learn, and perform actions.To address this problem, this dissertation models dimensions of physical causality associated with common action verbs. Based on such modeling, several approaches are developed to incorporate causality knowledge to language grounding, visual causality reasoning, and commonsense story comprehension.
Show less
- Title
- Finding optimized bounding boxes of polytopes in d-dimensional space and their properties in k-dimensional projections
- Creator
- Shahid, Salman (Of Michigan State University)
- Date
- 2014
- Collection
- Electronic Theses & Dissertations
- Description
-
Using minimal bounding boxes to encapsulate or approximate a set of points in d-dimensional space is a non-trivial problem that has applications in a variety of fields including collision detection, object rendering, high dimensional databases and statistical analysis to name a few. While a significant amount of work has been done on the three dimensional variant of the problem (i.e. finding the minimum volume bounding box of a set of points in three dimensions), it is difficult to find a...
Show moreUsing minimal bounding boxes to encapsulate or approximate a set of points in d-dimensional space is a non-trivial problem that has applications in a variety of fields including collision detection, object rendering, high dimensional databases and statistical analysis to name a few. While a significant amount of work has been done on the three dimensional variant of the problem (i.e. finding the minimum volume bounding box of a set of points in three dimensions), it is difficult to find a simple method to do the same for higher dimensions. Even in three dimensions existing methods suffer from either high time complexity or suboptimal results with a speed up in execution time. In this thesis we present a new approach to find the optimized minimum bounding boxes of a set of points defining convex polytopes in d-dimensional space. The solution also gives the optimal bounding box in three dimensions with a much simpler implementation while significantly speeding up the execution time for a large number of vertices. The basis of the proposed approach is a series of unique properties of the k-dimensional projections that are leveraged into an algorithm. This algorithm works by constructing the convex hulls of a given set of points and optimizing the projections of those hulls in two dimensional space using the new concept of Simultaneous Local Optimal. We show that the proposed algorithm provides significantly better performances than those of the current state of the art approach on the basis of time and accuracy. To illustrate the importance of the result in terms of a real world application, the optimized bounding box algorithm is used to develop a method for carrying out range queries in high dimensional databases. This method uses data transformation techniques in conjunction with a set of heuristics to provide significant performance improvement.
Show less
- Title
- Non-coding RNA identification in large-scale genomic data
- Creator
- Yuan, Cheng
- Date
- 2014
- Collection
- Electronic Theses & Dissertations
- Description
-
Noncoding RNAs (ncRNAs), which function directly as RNAs without translating into proteins, play diverse and important biological functions. ncRNAs function not only through their primary structures, but also secondary structures, which are defined by interactions between Watson-Crick and wobble base pairs. Common types of ncRNA include microRNA, rRNA, snoRNA, tRNA. Functions of ncRNAs vary among different types. Recent studies suggest the existence of large number of ncRNA genes....
Show moreNoncoding RNAs (ncRNAs), which function directly as RNAs without translating into proteins, play diverse and important biological functions. ncRNAs function not only through their primary structures, but also secondary structures, which are defined by interactions between Watson-Crick and wobble base pairs. Common types of ncRNA include microRNA, rRNA, snoRNA, tRNA. Functions of ncRNAs vary among different types. Recent studies suggest the existence of large number of ncRNA genes. Identification of novel and known ncRNAs becomes increasingly important in order to understand their functionalities and the underlying communities.Next-generation sequencing (NGS) technology sheds lights on more comprehensive and sensitive ncRNA annotation. Lowly transcribed ncRNAs or ncRNAs from rare species with low abundance may be identified via deep sequencing. However, there exist several challenges in ncRNA identification in large-scale genomic data. First, the massive volume of datasets could lead to very long computation time, making existing algorithms infeasible. Second, NGS has relatively high error rate, which could further complicate the problem. Third, high sequence similarity among related ncRNAs could make them difficult to identify, resulting in incorrect output. Fourth, while secondary structures should be adopted for accurate ncRNA identification, they usually incur high computational complexity. In particular, some ncRNAs contain pseudoknot structures, which cannot be effectively modeled by the state-of-the-art approach. As a result, ncRNAs containing pseudoknots are hard to annotate.In my PhD work, I aimed to tackle the above challenges in ncRNA identification. First, I designed a progressive search pipeline to identify ncRNAs containing pseudoknot structures. The algorithms are more efficient than the state-of-the-art approaches and can be used for large-scale data. Second, I designed a ncRNA classification tool for short reads in NGS data lacking quality reference genomes. The initial homology search phase significantly reduces size of the original input, making the tool feasible for large-scale data. Last, I focused on identifying 16S ribosomal RNAs from NGS data. 16S ribosomal RNAs are very important type of ncRNAs, which can be used for phylogenic study. A set of graph based assembly algorithms were applied to form longer or full-length 16S rRNA contigs. I utilized paired-end information in NGS data, so lowly abundant 16S genes can also be identified. To reduce the complexity of problem and make the tool practical for large-scale data, I designed a list of error correction and graph reduction techniques for graph simplification.
Show less
- Title
- Unobtrusive physiological monitoring using smartphones
- Creator
- Hao, Tian (Research scientist)
- Date
- 2015
- Collection
- Electronic Theses & Dissertations
- Description
-
"This thesis presents an in-depth investigation in unobtrusive smartphone-based physiological monitoring, which aims to help people get healthier and fitter in a more efficient and less costly way." -- Abstract.
- Title
- Image annotation and tag completion via kernel metric learning and noisy matrix recovery
- Creator
- Feng, Zheyun
- Date
- 2016
- Collection
- Electronic Theses & Dissertations
- Description
-
In the last several years, with the ever-growing popularity of digital photography and social media, the number of images with user-provided tags has increased enormously. Due to the large amount and content versatility of these images, there is an urgent need to categorize, index, retrieve and browse these images via semantic tags (also called attributes or keywords). Following this trend, image annotation or tag completion out of missing and noisy given tags over large scale datasets has...
Show moreIn the last several years, with the ever-growing popularity of digital photography and social media, the number of images with user-provided tags has increased enormously. Due to the large amount and content versatility of these images, there is an urgent need to categorize, index, retrieve and browse these images via semantic tags (also called attributes or keywords). Following this trend, image annotation or tag completion out of missing and noisy given tags over large scale datasets has become an extremely hot topic in the interdisciplinary areas of machine learning and computer vision.The overarching goal of this thesis is to reassess the image annotation and tag completion algorithms that mainly capture the essential relationship both between and within images and tags even when the given tag information is incomplete or noisy, so as to achieve a better performance in terms of both effectiveness and efficiency in image annotation and other tag relevant tasks including tag completion, tag ranking and tag refinement.One of the key challenges in search-based image annotation models is to define an appropriate similarity measure (distance metric) between images, so as to assign unlabeled images with tags that are shared among similar labeled training images. Many kernel metric learning (KML) algorithms have been developed to serve as such a nonlinear distance metric. However, most of them suffer from high computational cost since the learned kernel metric needs to be projected into a positive semi-definite (PSD) cone. Besides, in image annotation tasks, existing KML algorithms require to convert image annotation tags into binary constraints, which lead to a significant semantic information loss and severely reduces the annotation performance.In this dissertation we propose a robust kernel metric learning (RKML) algorithm based on regression technique that is able to directly utilize the image tags. RKML is computationally efficient since the PSD property is automatically ensured by the regression technique. Numeric constraints over tags are also applied to better exploit the tag information and hence improve the annotation accuracy. Further, theoretical guarantees for RKML are provided, and its efficiency and effectiveness are also verified empirically by comparing it to state-of-the-art approaches of both distance metric learning and image annotation.Since the user-provided image tags are always incomplete and noisy, we also propose a tag completion algorithm by noisy matrix recovery (TCMR) to simultaneously enrich the missing tags and remove the noisy ones. TCMR assumes that the observed tags are independently sampled from unknown distributions that are represented by a tag matrix, and our goal is to recover that tag matrix based on the partially revealed tags which could be noisy. We provide theoretical guarantees for TCMR with recovery error bounds. In addition, a graph Laplacian based component is introduced to enforce the recovered tags to be consistent with the visual contents of images. Our empirical study with multiple benchmark datasets for image tagging shows that the proposed algorithm outperforms state-of-the-art approaches in terms of both effectiveness and efficiency when handling missing and noisy tags.
Show less
- Title
- Structure and evolutionary dynamics in fitness landscapes
- Creator
- Pakanati, Anuraag R.
- Date
- 2015
- Collection
- Electronic Theses & Dissertations
- Description
-
Evolution can be conceptualized as an optimization algorithm that allows populations to search through genotypes for those that produce high fitness solutions. This search process is commonly depicted as exploring a “fitness landscape”, which combines similarity relationships among genotypes with the concept of a genotype-fitness map. As populations adapt to their fitness landscape, they accumulate information about the fitness landscape in which they live. A greater understanding of...
Show moreEvolution can be conceptualized as an optimization algorithm that allows populations to search through genotypes for those that produce high fitness solutions. This search process is commonly depicted as exploring a “fitness landscape”, which combines similarity relationships among genotypes with the concept of a genotype-fitness map. As populations adapt to their fitness landscape, they accumulate information about the fitness landscape in which they live. A greater understanding of evolution on fitness landscapes will help elucidate fundamental evolutionary processes. I examine methods of estimating information acquisition in evolving populations and find that these techniques have largely ignored the effects of common descent. Since information is estimated by measuring conserved genomic regions across a population, common descent can create a severe bias by increasing similarities among unselected regions. I introduce a correction method to compensate for the effects of common descent on genomic information and empirically demonstrate its efficacy.Next, I explore three instantiations of NK, Avida, and RNA fitness landscapes to better understand structural properties such as the distribution of peaks and the size of basins of attraction. I find that the fitness of peaks is correlated with the fitness of peaks within their neighborhood, and that the size of peaks' basins of attraction tends to be proportional to the heights of the peaks. Finally, I visualize local dynamics and perform a detailed comparison between the space of what evolutionary trajectories are technically possible from a single starting point and the results of actual evolving populations.
Show less
- Title
- Automated addition of fault-tolerance via lazy repair and graceful degradation
- Creator
- Lin, Yiyan
- Date
- 2015
- Collection
- Electronic Theses & Dissertations
- Description
-
In this dissertation, we concentrate on the problem of automated addition of fault-tolerance that transforms a fault-intolerant program to be a fault-tolerant program. We solve this problem via model repair. Model repair is a correct-by-construct technique to revise an existing model so that the revised model satisfies the given correctness criteria, such as safety, liveness, or fault-tolerance. We consider two problems of using model repair to add fault-tolerance. First, if the repaired...
Show moreIn this dissertation, we concentrate on the problem of automated addition of fault-tolerance that transforms a fault-intolerant program to be a fault-tolerant program. We solve this problem via model repair. Model repair is a correct-by-construct technique to revise an existing model so that the revised model satisfies the given correctness criteria, such as safety, liveness, or fault-tolerance. We consider two problems of using model repair to add fault-tolerance. First, if the repaired model violates the assumptions (e.g., partial observability, inability to detect crashed processes, etc) made in the underlying system, then it cannot be implemented. We denote these requirements as realizability constraints. Second, the addition of fault-tolerance may fail if the program cannot fully recover after certain faults occur. In this dissertation, we propose a lazy repair approach to address realizability issues in adding fault-tolerance. Additionally, we propose a technique to automatically add graceful degradation to a program, so that the program can recover with partial functionality (that is identified by the designer to be the critical functionality) if full recovery is impossible.A model repair technique transforms a model to another model that satisfies a new set of properties. Such a transformation should also maintain the mapping between the model and the underlying program. For example, in a distributed program, every process is restricted to read (or write) some variables in other processes. A model that represents this program should also disallow the process to read (or write) those inaccessable variables. If these constraints are violated, then the corresponding model will be unrealizable. An unrealizable model (in this context, a model that violates the read/write restrictions) may make it impossible to obtain the corresponding implementation.%In this dissertation, we call the read (or write) restriction as a realizability constraint in distributed systems. An unrealizable model (a model that violates the realizability constraints) may complicate the implementation by introducing extra amount of modification to the program. Such modification may in turn break the program's correctness.Resolving realizability constraints increases the complexity of model repair. Existing model repair techniques introduce heuristics to reduce the complexity. However, this heuristic-based approach is designed and optimized specifically for distributed programs. We need a more generic model repair approach for other types of programs, e.g., synchronous programs, cyber-physical programs, etc. Hence, in this dissertation, we propose a model repair technique, i.e., lazy repair, to add fault-tolerance to programs with different types of realizability constraints. It involves two steps. First, we only focus on repairing to obtain a model that satisfies correctness criteria while ignoring realizability constraints. In the second step, we repair this model further by removing behaviors while ensuring that the desired specification is preserved. The lazy repair approach simplifies the process of developing heuristics, and provides a tradeoff in terms of the time saved in the first step and the extra work required in the second step. We demonstrate that lazy repair is applicable in the context of distributed systems, synchronous systems and cyber-physical systems.In addition, safety critical systems such as airplanes, automobiles and elevators should operate with high dependability in the presence of faults. If the occurrence of faults breaks down some components, the system may not be able to fully recover. In this scenario, the system can still operate with remaining resources and deliver partial but core functionality, i.e., to display graceful degradation. Existing model repair approaches, such as addition of fault-tolerance, cannot transform a program to provide graceful degradation. In this dissertation, we propose a technique to add fault-tolerance to a program with graceful degradation. In the absence of faults, such a program exhibits ideal behaviors. In the presence of faults, the program is allowed to recover with reduced functionality. This technique involves two steps. First, it automatically generates a program with graceful degradation based on the input fault-intolerant program. Second, it adds fault-tolerance to the output program from first step. We demonstrate that this technique is applicable in the context of high atomicity programs as well as low atomicity programs (i.e., distributed programs). We also present a case study on adding multi-graceful degradation to a dangerous gas detection and ventilation system. Through this case study, we show that our approach can assist the designer to obtain a program that behaves like the deployed system.
Show less
- Title
- Computational identification and analysis of non-coding RNAs in large-scale biological data
- Creator
- Lei, Jikai
- Date
- 2015
- Collection
- Electronic Theses & Dissertations
- Description
-
Non-protein-coding RNAs (ncRNAs) are RNA molecules that function directly at the level of RNA without translating into protein. They play important biological functions in all three domains of life, i.e. Eukarya, Bacteria and Archaea. To understand the working mechanisms and the functions of ncRNAs in various species, a fundamental step is to identify both known and novel ncRNAs from large-scale biological data.Large-scale genomic data includes both genomic sequence data and NGS sequencing...
Show moreNon-protein-coding RNAs (ncRNAs) are RNA molecules that function directly at the level of RNA without translating into protein. They play important biological functions in all three domains of life, i.e. Eukarya, Bacteria and Archaea. To understand the working mechanisms and the functions of ncRNAs in various species, a fundamental step is to identify both known and novel ncRNAs from large-scale biological data.Large-scale genomic data includes both genomic sequence data and NGS sequencing data. Both types of genomic data provide great opportunity for identifying ncRNAs. For genomic sequence data, a lot of ncRNA identification tools that use comparative sequence analysis have been developed. These methods work well for ncRNAs that have strong sequence similarity. However, they are not well-suited for detecting ncRNAs that are remotely homologous. Next generation sequencing (NGS), while it opens a new horizon for annotating and understanding known and novel ncRNAs, also introduces many challenges. First, existing genomic sequence searching tools can not be readily applied to NGS data because NGS technology produces short, fragmentary reads. Second, most NGS data sets are large-scale. Existing algorithms are infeasible on NGS data because of high resource requirements. Third, metagenomic sequencing, which utilizes NGS technology to sequence uncultured, complex microbial communities directly from their natural inhabitants, further aggravates the difficulties. Thus, massive amount of genomic sequence data and NGS data calls for efficient algorithms and tools for ncRNA annotation.In this dissertation, I present three computational methods and tools to efficiently identify ncRNAs from large-scale biological data. Chain-RNA is a tool that combines both sequence similarity and structure similarity to locate cross-species conserved RNA elements with low sequence similarity in genomic sequence data. It can achieve significantly higher sensitivity in identifying remotely conserved ncRNA elements than sequence based methods such as BLAST, and is much faster than existing structural alignment tools. miR-PREFeR (miRNA PREdiction From small RNA-Seq data) utilizes expression patterns of miRNA and follows the criteria for plant microRNA annotation to accurately predict plant miRNAs from one or more small RNA-Seq data samples. It is sensitive, accurate, fast and has low-memory footprint. metaCRISPR focuses on identifying Clustered Regularly Interspaced Short Palindromic Repeats (CRISPRs) from large-scale metagenomic sequencing data. It uses a kmer hash table to efficiently detect reads that belong to CRISPRs from the raw metagonmic data set. Overlap graph based clustering is then conducted on the reduced data set to separate different CRSIPRs. A set of graph based algorithms are used to assemble and recover CRISPRs from the clusters.
Show less
- Title
- Harnessing low-pass filter defects for improving wireless link performance : measurements and applications
- Creator
- Renani, Alireza Ameli
- Date
- 2018
- Collection
- Electronic Theses & Dissertations
- Description
-
"The design trade-offs of transceiver hardware are crucial to the performance of wireless systems. The effect of such trade-offs on individual analog and digital components are vigorously studied, but their systemic impacts beyond component-level remain largely unexplored. In this dissertation, we present an in-depth study to characterize the surprisingly notable systemic impacts of low-pass filter design, which is a small yet indispensable component used for shaping spectrum and rejecting...
Show more"The design trade-offs of transceiver hardware are crucial to the performance of wireless systems. The effect of such trade-offs on individual analog and digital components are vigorously studied, but their systemic impacts beyond component-level remain largely unexplored. In this dissertation, we present an in-depth study to characterize the surprisingly notable systemic impacts of low-pass filter design, which is a small yet indispensable component used for shaping spectrum and rejecting interference. Using a bottom-up approach, we examine how signal-level distortions caused by the trade-offs of low-pass filter design propagate to the upper-layers of wireless communication, reshaping bit error patterns and degrading link performance of today's 802.11 systems. Moreover, we propose a novel unequal error protection algorithm that harnesses low-pass filter defects for improving wireless LAN throughput, particularly to be used in forward error correction, channel coding, and applications such as video streaming. Lastly, we conduct experiments to evaluate the unequal error protection algorithm in video streaming, and we present substantial enhancements of video quality in mobile environments."--Page ii.
Show less
- Title
- Smartphone-based sensing systems for data-intensive applications
- Creator
- Moazzami, Mohammad-Mahdi
- Date
- 2017
- Collection
- Electronic Theses & Dissertations
- Description
-
"Supported by advanced sensing capabilities, increasing computational resources and the advances in Artificial Intelligence, smartphones have become our virtual companions in our daily life. An average modern smartphone is capable of handling a wide range of tasks including navigation, advanced image processing, speech processing, cross app data processing and etc. The key facet that is common in all of these applications is the data intensive computation. In this dissertation we have taken...
Show more"Supported by advanced sensing capabilities, increasing computational resources and the advances in Artificial Intelligence, smartphones have become our virtual companions in our daily life. An average modern smartphone is capable of handling a wide range of tasks including navigation, advanced image processing, speech processing, cross app data processing and etc. The key facet that is common in all of these applications is the data intensive computation. In this dissertation we have taken steps towards the realization of the vision that makes the smartphone truly a platform for data intensive computations by proposing frameworks, applications and algorithmic solutions. We followed a data-driven approach to the system design. To this end, several challenges must be addressed before smartphones can be used as a system platform for data-intensive applications. The major challenge addressed in this dissertation include high power consumption, high computation cost in advance machine learning algorithms, lack of real-time functionalities, lack of embedded programming support, heterogeneity in the apps, communication interfaces and lack of customized data processing libraries. The contribution of this dissertation can be summarized as follows. We present the design, implementation and evaluation of the ORBIT framework, which represents the first system that combines the design requirements of a machine learning system and sensing system together at the same time. We ported for the first time off-the-shelf machine learning algorithms for real-time sensor data processing to smartphone devices. We highlighted how machine learning on smartphones comes with severe costs that need to be mitigated in order to make smartphones capable of real-time data-intensive processing. From application perspective we present SPOT. SPOT aims to address some of the challenges discovered in mobile-based smart-home systems. These challenges prevent us from achieving the promises of smart-homes due to heterogeneity in different aspects of smart devices and the underlining systems. We face the following major heterogeneities in building smart-homes:: (i) Diverse appliance control apps (ii) Communication interface, (iii) Programming abstraction. SPOT makes the heterogeneous characteristics of smart appliances transparent, and by that it minimizes the burden of home automation application developers and the efforts of users who would otherwise have to deal with appliance-specific apps and control interfaces. From algorithmic perspective we introduce two systems in the smartphone-based deep learning area: Deep-Crowd-Label and Deep-Partition. Deep neural models are both computationally and memory intensive, making them difficult to deploy on mobile applications with limited hardware resources. On the other hand, they are the most advanced machine learning algorithms suitable for real-time sensing applications used in the wild. Deep-Partition is an optimization-based partitioning meta-algorithm featuring a tiered architecture for smartphone and the back-end cloud. Deep-Partition provides a profile-based model partitioning allowing it to intelligently execute the Deep Learning algorithms among the tiers to minimize the smartphone power consumption by minimizing the deep models feed-forward latency. Deep-Crowd-Label is prototyped for semantically labeling user's location. It is a crowd-assisted algorithm that uses crowd-sourcing in both training and inference time. It builds deep convolutional neural models using crowd-sensed images to detect the context (label) of indoor locations. It features domain adaptation and model extension via transfer learning to efficiently build deep models for image labeling. The work presented in this dissertation covers three major facets of data-driven and compute-intensive smartphone-based systems: platforms, applications and algorithms; and helps to spurs new areas of research and opens up new directions in mobile computing research."--Pages ii-iii.
Show less
- Title
- Fluid animation on deforming surface meshes
- Creator
- Wang, Xiaojun (Graduate of Michigan State University)
- Date
- 2017
- Collection
- Electronic Theses & Dissertations
- Description
-
"We explore methods for visually plausible fluid simulation on deforming surfaces with inhomogeneous diffusion properties. While there are methods for fluid simulation on surfaces, not much research effort focused on the influence of the motion of underlying surface, in particular when it is not a rigid surface, such as knitted or woven textiles in motion. The complexity involved makes the simulation challenging to account for the non-inertial local frames typically used to describe the...
Show more"We explore methods for visually plausible fluid simulation on deforming surfaces with inhomogeneous diffusion properties. While there are methods for fluid simulation on surfaces, not much research effort focused on the influence of the motion of underlying surface, in particular when it is not a rigid surface, such as knitted or woven textiles in motion. The complexity involved makes the simulation challenging to account for the non-inertial local frames typically used to describe the motion and the anisotropic effects in diffusion, absorption, adsorption. Thus, our primary goal is to enable fast and stable method for such scenarios. First, in preparation of the material properties for the surface domain, we describe textiles with salient feature direction by bulk material property tensors in order to reduce the complexity, by employing 2D homogenization technique, which effectively turns microscale inhomogeneous properties into homogeneous properties in macroscale descriptions. We then use standard texture mapping techniques to map these tensors to triangles in the curved surface mesh, taking into account the alignment of each local tangent space with correct feature directions of the macroscale tensor. We show that this homogenization tool is intuitive, flexible and easily adjusted. Second, for efficient description of the deforming surface, we offer a new geometry representation for the surface with solely angles instead of vertex coordinates, to reduce storage for the motion of underlying surface. Since our simulation tool relies heavily on long sequences of 3D curved triangular meshes, it is worthwhile exploring such efficient representations to make our tool practical by reducing the memory access during real-time simulations as well as reducing the file sizes. Inspired by angle-based representations for tetrahedral meshes, we use spectral method to restore curved surface using both angles of the triangles and dihedral angles between adjacent triangles in the mesh. Moreover, in many surface deformation sequences, it is often sufficient to update the dihedral angles while keeping the triangle interior angles fixed. Third, we propose a framework for simulating various effects of fluid flowing on deforming surfaces. We directly applied our simulator on curved surface meshes instead of in parameter domains, whereas many existing simulation methods require a parameterization on the surface. We further demonstrate that fictitious forces induced by the surface motion can be added to the surface-based simulation at a small additional cost. These fictitious forces can be decomposed into different components. Only the rectilinear and Coriolis components are relevant to our choice of local frames. Other effects, such as diffusion, adsorption, absorption, and evaporation are also incorporated for realistic stain simulation. Finally, we explore the extraction of Lagrangian Coherent Structure (LCS), which is often referred to as the skeleton of fluid motion. The LCS structures are often described by ridges of the finite time Lyapunov exponent (FTLE) fields, which describe the extremal stretching of fluid parcels following the flow. We proposed a novel improvement to the ridge marching algorithm, which extract such ridges robustly for the typically noisy FTLE estimates even in well-defined fluid flows. Our results are potentially applicable to visualizing and controlling fluid trajectory patterns. In contrast to current methods for LCS calculation, which are only applicable to flat 2D or 3D domains and sensitive to noise, our ridge extraction is readily applicable to curved surfaces even when they are deforming. The collection of these computational tools will facilitate generation of realistic and easy to adjust surface fluid animation with various physically plausible effects on surface."--Pages ii-iii.
Show less
- Title
- Data clustering with pairwise constraints
- Creator
- Yi, Jinfeng
- Date
- 2014
- Collection
- Electronic Theses & Dissertations
- Description
-
The classical unsupervised clustering is an ill-posed problem due to the absence of a unique clustering criteria. This issue can be addressed by introducing additional supervised information, usually casts in the form of pairwise constraints, to the clustering procedure. Depending on the sources, most pairwise constraints can be classified into two categories: (i) pairwise constraints collected from a set of non-expert crowd workers, which leads to the problem of crowdclustering, and (ii)...
Show moreThe classical unsupervised clustering is an ill-posed problem due to the absence of a unique clustering criteria. This issue can be addressed by introducing additional supervised information, usually casts in the form of pairwise constraints, to the clustering procedure. Depending on the sources, most pairwise constraints can be classified into two categories: (i) pairwise constraints collected from a set of non-expert crowd workers, which leads to the problem of crowdclustering, and (ii) pairwise constraints collected from oracle or experts, which leads to the problem of semi-supervised clustering. In both cases, the costs of collecting pairwise constraints can be expensive, thus it is important to identify the minimal number of pairwise constraints needed to accurately recover the underlying true data partition, also known as a sample complexity problem.In this thesis, we first analyze the sample complexity of crowdclustering. At first, we propose a novel crowdclustering approach based on the theory of matrix completion. Unlike the existing crowdclustering algorithm that is based on a Bayesian generative model, the proposed approach is more desirable since it only needs a much less number of crowdsourced pairwise annotations to accurately cluster all the objects. Our theoretical analysis shows that in order to accurately cluster $N$ objects, only $O(N\log^2 N)$ randomly sampled pairs should be annotated by crowd workers. To further reduce the sample complexity, we then introduce a semi-crowdsourced clustering framework that is able to effectively incorporate the low-level features of the objects to be clustered. In this framework, we only need to sample a subset of $n \ll N$ objects and generate their pairwise constraints via crowdsourcing. After completing a $n \times n$ similarity matrix using the proposed crowdclustering algorithm, we can further recover a $N \times N$ similarity matrix by applying a regression-based distance metric learning algorithm to the completed smaller size similarity matrix. This enables us to reliably cluster $N$ objects with only $O(n\log^2 n)$ crowdsourced pairwise constraints.Next, we study the problem of sample complexity in semi-supervised clustering. To this end, we propose a novel convex semi-supervised clustering approach based on the theory of matrix completion. In order to reduce the number of pairwise constraints needed %to achieve a perfect data partitioning,we apply a nature assumption that the feature representationsof the objects are able to reflect the similarities between objects. This enables us to only utilize $O(\log N)$ pairwiseconstraints to perfectly recover the data partition of $N$ objects.Lastly, in addition to sample complexity that relates to labeling costs, we also consider the computational costs of semi-supervised clustering.%In addition to sample complexity that relates to the labeling costs, we also consider the computational cost of semi-supervised clustering in the final part of this thesis.Specifically, we study the problem of efficiently updating clustering results when the pairwise constraints are generated sequentially, a common case in various real-world applications such as social networks. To address this issue, we develop a dynamic semi-supervised clustering algorithm that casts the clustering problem into a searching problem in a feasibleconvex space, i.e., a convex hull with its extreme points being an ensemble of multiple data partitions. Unlike classical semi-supervised clustering algorithms that need to re-optimize their objective functions when new pairwise constraints are generated, the proposed method only needs to update a low-dimensional vector and its time complexity is irrelevant to the number of data points to be clustered. This enables us to update large-scale clustering results in an extremely efficient way.
Show less
- Title
- Large-scale high dimensional distance metric learning and its application to computer vision
- Creator
- Qian, Qi
- Date
- 2015
- Collection
- Electronic Theses & Dissertations
- Description
-
Learning an appropriate distance function (i.e., similarity) is one of the key tasks in machine learning, especially for distance based machine learning algorithms, e.g., $k$-nearest neighbor classifier, $k$-means clustering, etc. Distance metric learning (DML), the subject to be studied in this dissertation, is designed to learn a metric that pulls the examples from the same class together and pushes the examples from different classes away from each other. Although many DML algorithms have...
Show moreLearning an appropriate distance function (i.e., similarity) is one of the key tasks in machine learning, especially for distance based machine learning algorithms, e.g., $k$-nearest neighbor classifier, $k$-means clustering, etc. Distance metric learning (DML), the subject to be studied in this dissertation, is designed to learn a metric that pulls the examples from the same class together and pushes the examples from different classes away from each other. Although many DML algorithms have been developed in the past decade, most of them can handle only small data sets with hundreds of features, significantly limiting their applications to real world applications that often involve millions of training examples represented by hundreds of thousands of features. Three main challenges are encountered to learn the metric from these large-scale high dimensional data: (i) To make sure that the learned metric is a Positive Semi-Definitive (PSD) matrix, a projection into the PSD cone is required at every iteration, whose cost is cubic in the dimensionality making it unsuitable for high dimensional data; (ii) The number of variables that needs to be optimized in DML is quadratic in the dimensionality, which results in the slow convergence rate in optimization and high requirement of memory storage; (iii) The number of constraints used by DML is at least quadratic, if not cubic, in the number of examples depending on if pairwise constraints or triplet constraints are used in DML. Besides, features can be redundant due to high dimensional representations (e.g., face features) and DML with feature selection is preferred for these applications.The main contribution of this dissertation is to address these challenges both theoretically and empirically. First, for the challenge arising from the PSD projection, we exploit the mini-batch strategy and adaptive sampling with smooth loss function to significantly reduce the number of updates (i.e., projections) while keeping the similar performance. Second, for the challenge arising from high dimensionality, we propose a dual random projection approach, which enjoys the light computation due to the usage of random projection and at the same time, significantly improves the effectiveness of random projection. Third, for the challenge with large-scale constraints, we develop a novel multi-stage metric learning framework. It divides the original optimization problem into multiple stages. It reduces the computation by adaptively sampling a small subset of constraints at each stage. Finally, to handle redundant features with group property, we develop a greedy algorithm that selects feature group and learns the corresponding metric simultaneously at each iteration leading to further improvement of learning efficiency when combined with adaptive mini-batch strategy and incremental sampling. Besides the theoretical and empirical investigation of DML on the benchmark datasets of machine learning, we also apply the proposed methods to several important computer vision applications (i.e., fine-grained visual categorization (FGVC) and face recognition).
Show less
- Title
- Unconstrained 3D face reconstruction from photo collections
- Creator
- Roth, Joseph (Software engineer)
- Date
- 2016
- Collection
- Electronic Theses & Dissertations
- Description
-
This thesis presents a novel approach for 3D face reconstruction from unconstrained photo collections. An unconstrained photo collection is a set of face images captured under an unknown and diverse variation of poses, expressions, and illuminations. The output of the proposed algorithm is a true 3D face surface model represented as a watertight triangulated surface with albedo data colloquially referred to as texture information. Reconstructing a 3D understanding of a face based on 2D input...
Show moreThis thesis presents a novel approach for 3D face reconstruction from unconstrained photo collections. An unconstrained photo collection is a set of face images captured under an unknown and diverse variation of poses, expressions, and illuminations. The output of the proposed algorithm is a true 3D face surface model represented as a watertight triangulated surface with albedo data colloquially referred to as texture information. Reconstructing a 3D understanding of a face based on 2D input is a long-standing computer vision problem. Traditional photometric stereo-based reconstruction techniques work on aligned 2D images and produce a 2.5D depth map reconstruction. We extend face reconstruction to work with a true 3D model, allowing us to enjoy the benefits of using images from all poses, up to and including profiles. To use a 3D model, we propose a novel normal field-based Laplace editing technique which allows us to deform a triangulated mesh to match the observed surface normals. Unlike prior work that require large photo collections, we formulate an approach to adapt to photo collections with few images of potentially poor quality. We achieve this through incorporating prior knowledge about face shape by fitting a 3D Morphable Model to form a personalized template before using a novel analysis-by-synthesis photometric stereo formulation to complete the fine face details. A structural similarity-based quality measure allows evaluation in the absence of ground truth 3D scans. Superior large-scale experimental results are reported on Internet, synthetic, and personal photo collections.
Show less
- Title
- Novel learning algorithms for mining geospatial data
- Creator
- Yuan, Shuai (Software engineer)
- Date
- 2017
- Collection
- Electronic Theses & Dissertations
- Description
-
Geospatial data have a wide range of applicability in many disciplines, including environmental science, urban planning, healthcare, and public administration. The proliferation of such data in recent years have presented opportunities to develop novel data mining algorithms for modeling and extracting useful patterns from the data. However, there are many practical issues remain that must be addressed before the algorithms can be successfully applied to real-world problems. First, the...
Show moreGeospatial data have a wide range of applicability in many disciplines, including environmental science, urban planning, healthcare, and public administration. The proliferation of such data in recent years have presented opportunities to develop novel data mining algorithms for modeling and extracting useful patterns from the data. However, there are many practical issues remain that must be addressed before the algorithms can be successfully applied to real-world problems. First, the algorithms must be able to incorporate spatial relationships and other domain constraints defined by the problem. Second, the algorithms must be able to handle missing values, which are common in many geospatial data sets. In particular, the models constructed by the algorithms may need to be extrapolated to locations with no observation data. Another challenge is to adequately capture the nonlinear relationship between the predictor and response variables of the geospatial data. Accurate modeling of such relationship is not only a challenge, it is also computationally expensive. Finally, the variables may interact at different spatial scales, making it necessary to develop models that can handle multi-scale relationships present in the geospatial data. This thesis presents the novel algorithms I have developed to overcome the practical challenges of applying data mining to geospatial datasets. Specifically, the algorithms will be applied to both supervised and unsupervised learning problems such as cluster analysis and spatial prediction. While the algorithms are mostly evaluated on datasets from the ecology domain, they are generally applicable to other geospatial datasets with similar characteristics. First, a spatially constrained spectral clustering algorithm is developed for geospatial data. The algorithm provides a flexible way to incorporate spatial constraints into the spectral clustering formulation in order to create regions that are spatially contiguous and homogeneous. It can also be extended to a hierarchical clustering setting, enabling the creation of fine-scale regions that are nested wholly within broader-scale regions. Experimental results suggest that the nested regions created using the proposed approach are more balanced in terms of their sizes compared to the regions found using traditional hierarchical clustering methods. Second, a supervised hash-based feature learning algorithm is proposed for modeling nonlinear relationships in incomplete geospatial data. The proposed algorithm can simultaneously infer missing values while learning a small set of discriminative, nonlinear features of the geospatial data. The efficacy of the algorithm is demonstrated using synthetic and real-world datasets. Empirical results show that the algorithm is more effective than the standard approach of imputing the missing values before applying nonlinear feature learning in more than 75% of the datasets evaluated in the study. Third, a multi-task learning framework is developed for modeling multiple response variables in geospatial data. Instead of training the local models independently for each response variable at each location, the framework simultaneously fits the local models for all response variables by optimizing a joint objective function with trace-norm regularization. The framework also leverages the spatial autocorrelation between locations as well as the inherent correlation between response variables to improve prediction accuracy. Finally, a multi-level, multi-task learning framework is proposed to effectively train predictive models from nested geospatial data containing predictor variables measured at multiple spatial scales. The framework enables distinct models to be developed for each coarse- scale region using both its fine-level and coarse-level features. It also allows information to be shared among the models through a common set of latent features. Empirical results show that such information sharing helps to create more robust models especially for regions with limited or no training data. Another advantage of using the multi-level, multi-task learning framework is that it can automatically identify potential cross-scale interactions between the regional and local variables.
Show less
- Title
- TEACHERS IN SOCIAL MEDIA : A DATA SCIENCE PERSPECTIVE
- Creator
- Karimi, Hamid
- Date
- 2021
- Collection
- Electronic Theses & Dissertations
- Description
-
Social media has become an integral part of human life in the 21st century. The number of social media users was estimated to be around 3.6 billion individuals in 2020. Social media platforms (e.g., Facebook) have facilitated interpersonal communication, diffusion of information, the creation of groups and communities, to name a few. As far as education systems are concerned, online social media has transformed and connected traditional social networks within the schoolhouse to a broader and...
Show moreSocial media has become an integral part of human life in the 21st century. The number of social media users was estimated to be around 3.6 billion individuals in 2020. Social media platforms (e.g., Facebook) have facilitated interpersonal communication, diffusion of information, the creation of groups and communities, to name a few. As far as education systems are concerned, online social media has transformed and connected traditional social networks within the schoolhouse to a broader and expanded world outside. In such an expanded virtual space, teachers engage in various activities within their communities, e.g., exchanging instructional resources, seeking new teaching methods, engaging in online discussions. Therefore, given the importance of teachers in social media and its tremendous impact on PK-12 education, in this dissertation, we investigate teachers in social media from a data science perspective. Our investigation in this direction is essentially an interdisciplinary endeavor bridging modern data science and education. In particular, we have made three contributions, as briefly discussed in the following. Current teachers in social media studies suffice to a small number of surveyed teachers while thousands of other teachers are on social media. This hinders us from conducting large-scale data-driven studies pertinent to teachers in social media. Aiming to overcome this challenge and further facilitate data-driven studies related to teachers in social media, we propose a novel method that automatically identifies teachers on Pinterest, an image-based social media popular among teachers. In this framework, we formulate the teacher identification problem as a positive unlabelled (PU) learning where positive samples are surveyed teachers, and unlabelled samples are their online friends. Using our framework, we build the largest dataset of teachers on Pinterest. With this dataset at our disposal, we perform an exploratory analysis of teachers on Pinterest while considering their genders. Our analysis incorporates two crucial aspects of teachers in social media. First, we investigate various online activities of male and female teachers, e.g., topics and sources of their curated resources, the professional language employed to describe their resources. Second, we investigate male and female teachers in the context of the social network (the graph) they belong to, e.g., structural centrality, gender homophily. Our analysis and findings in this part of the dissertation can serve as a valuable reference for many entities concerned with teachers' gender, e.g., principals, state, and federal governments.Finally, in the third part of the dissertation, we shed light on the diffusion of teacher-curated resources on Pinterest. First, we introduce three measures to characterize the diffusion process. Then, we investigate these three measures while considering two crucial characteristics of a resource, e.g., the topic and the source. Ultimately, we investigate how teacher attributes (e.g., the number of friends) affect the diffusion of their resources. The conducted diffusion analysis is the first of its kind and offers a deeper understating of the complex mechanism driving the diffusion of resources curated by teachers on Pinterest.
Show less
- Title
- Optimizing for Mental Representations in the Evolution of Artificial Cognitive Systems
- Creator
- Kirkpatrick, Douglas Andrew
- Date
- 2021
- Collection
- Electronic Theses & Dissertations
- Description
-
Mental representations, or sensor-independent internal models of the environment, are used to interpret the world and make decisions based upon that understanding. For example, a human sees dark clouds in the sky, recalls that often dark clouds mean rain (a mental representation), and consequently decides to wear a raincoat. I seek to identify, understand, and encourage the evolution of these representations in silico. Previous work identified an information-theoretic tool, referred to as R,...
Show moreMental representations, or sensor-independent internal models of the environment, are used to interpret the world and make decisions based upon that understanding. For example, a human sees dark clouds in the sky, recalls that often dark clouds mean rain (a mental representation), and consequently decides to wear a raincoat. I seek to identify, understand, and encourage the evolution of these representations in silico. Previous work identified an information-theoretic tool, referred to as R, that measures mental representations in artificial cognitive systems (e.g., Markov Brains or Recurrent Neural Networks). Further work found that selecting for R, along with task performance, in the evolution of artificial cognitive systems leads to better overall performance on a given task. Here I explore the implications and opportunities of this modified selection process, referred to as R-augmentation. After an overview of common methods, techniques, and computational substrates in Chapter 1, a series of working chapters experimentally demonstrate the capabilities and possibilities of R-augmentation. First, in Chapter 2, I address concerns regarding potential limitations of R-augmentation. This includes an refutation of suspected negative impacts on the system’s ability to generalize within-domain and the system’s robustness to sensor noise. To the contrary, the systems evolved with R-augmentation tend to perform better than those evolved without, in the context of noisy environments and different computational components. In Chapter 3 I examine how R-augmentation works across different cognitive structures, focusing on the evolution of genetic programming related structures and the effect that augmentation has on the distribution of their representations. For Chapter 4, in the context of the all-component Markov Brain (referred to as a Buffet Brain, see [Hintze et al., 2019]) I analyze potential reasons that explain why R-augmentation works; the mechanism seems to be based on evolutionary dynamics as opposed to structural or component differences. Next, I demonstrate a novel usage of R-augmentation in Chapter 5; with R-augmentation, one can use far fewer training examples during evolution and the resulting systems still perform approximately as well as those that were trained on the full set of examples. This advantage in increased performance at low sample size is found in some examples of in-domain and out-domain generalization, with the ”worst-case” scenario being that the networks created by R-augmentation perform as well as their unaugmented equivalents. Lastly, in Chapter 6 I move beyond R-augmentation to explore using other neuro-correlates - particularly the distribution of representations, called smearedness - as part of the fitness function. I investigate the possibility of using MAP-Elites to identify an optimal value of smearedness for augmentation or for use as an optimization method in its own right. Taken together, these investigations demonstrate both the capabilities and limitations of R-augmentation, and open up pathways for future research.
Show less