Search results

Title: ACTIVE INFERENCE FOR PREDICTIVE MODELS OF SPATIO-TEMPORAL DOMAINS
Creator: Komurlu, Caner
Date: 2019
Description: Active inference is the method of selective information gathering during prediction in order to increase a predictive machine learning model's...
Show moreActive inference is the method of selective information gathering during prediction in order to increase a predictive machine learning model's prediction performance. Unlike active learning, active inference does not update the model, but rather provides the model with useful information during prediction to boost the prediction performance. To be able to work with active inference, a predictive model needs to exploit correlations among variables that need to be predicted. Then the model, while being provided with true values for some of the variables, can make more accurate predictions for the remaining variables.In this dissertation, I propose active inference methods for predictive models of spatio-temporal domains. I formulate and investigate active inference in two different domains: tissue engineering and wireless sensor networks. I develop active inference for dynamic Bayesian networks (DBNs) and feed-forward neural networks (FFNNs).First, I explore the effect of active inference in the tissue engineering domain. I design a dynamic Bayesian network (DBN) model for vascularization of a tissue development site. The DBN model predicts probabilities of blood vessel invasion in regional scale through time. Then utilizing spatio-temporal correlations between regions represented as variables in the DBN model, I develop an active inference technique to detect the optimal time to stop a wet lab experiment. The empirical study shows that the active inference is able to detect the optimal time and the results are coherent with domain simulations and lab experiments.In the second phase of my research, I develop variance-based active inference techniques for dynamic Bayesian networks for the purpose of battery saving for wireless sensor networks (WSN). I propose the expected variance reduction active inference method to detect variables that reduce the overall variance the most. I first propose a DBN model of a WSN. I then compare the prediction performance of the DBN with Gaussian processes and linear chain graphical models on three different WSN data using several baseline active inference methods. After showing that DBNs perform better than the baseline predictive models, I compare the performance of expected variance reduction active inference method with the performances of baseline methods on the DBN, and show the superiority of the expected variance reduction on the three WSN data sets.Finally, to address the inference complexity and the limitation of representing linear correlations due to Gaussian assumption, I replace the DBN representation with a feed-forward neural network (FFNN) model. I first explore techniques to integrate observed values into predictions on neural networks. I adopt the input optimization technique. Finally, I discover two problems: model error and optimization overfitting. I show that the input optimization can mitigate the model error. Lastly, I propose a validation-based regularization approach to solve the overfitting problem.
Show less

Title: DAMAGE ASSESSMENT OF CIVIL STRUCTURES AFTER NATURAL DISASTERS USING DEEP LEARNING AND SATELLITE IMAGERY
Creator: Jones, Scott F
Date: 2019
Description: Since 1980, millions of people have been harmed by natural disasters that have cost communities across the world over three trillion dollars....
Show moreSince 1980, millions of people have been harmed by natural disasters that have cost communities across the world over three trillion dollars. After a natural disaster has occurred, the creation of maps that identify the damage to buildings and infrastructure is imperative. Currently, many organizations perform this task manually, using pre- and post-disaster images and well-trained professionals to determine the degree and extent of damage. This manual task can take days to complete. I propose to do this task automatically using post-disaster satellite imagery. I use a pre-trained neural network, SegNet, and replaced its last layer with a simple damage classification scheme. This final layer of the network is re-trained using cropped segments of the satellite image of the disaster. The data were obtained from a publicly accessible source, the Copernicus EMS system. They provided three channel (RGB) reference and damage grading maps that were used to create the images of the ground truth and the damaged terrain. I then retrained the final layer of the network to identify civil structures that had been damaged. The resulting network was 85% accurate at labelling the pixels in an image of the disaster from typhoon Haiyan. The test results show that it is possible to create these maps quickly and efficiently.
Show less

Title: Fast mesh based reconstruction for cardiac-gated SPECT and methodology for medical image quality assessment
Creator: Massanes Basi, Francesc
Date: 2018
Description: In this work, we are studying two different subjects that are intricately connected. For the first subject we are considering tools to...
Show moreIn this work, we are studying two different subjects that are intricately connected. For the first subject we are considering tools to improve the quality of single photon emission computed tomography (SPECT) imaging. Currently, SPECT images assist physicians to evaluate perfusion levels within the myocardium, aide in the diagnosis of various types of carcinomas, and measure pulmonary function. The SPECT technique relies on injecting a radioactive material into the patient's body and then detecting the emitted radiation by means of a gamma camera. However, the amount of radioactive material that can be given to a patient is limited by the negative effects that the radiation will have on the patient's health. This causes SPECT images to be highly corrupted by noise. We will focus our work on cardiac SPECT, which adds the challenge of the heart's continuous motion during the acquisition process. First, we describe the methodology used in SPECT imaging and reconstruction. Our methodology uses a content adaptive model, which uses more samples on the regions of the body that we want to be reconstructed more accurately and less in other areas. Then we describe our algorithm and our novel implementation that lets us use the content adaptive model to perform the reconstruction. In this work, we show that our implementation outperforms the reconstruction method used for clinical applications. In the second subject we are evaluating tools to measure image quality in the context of medical diagnosis. In signal processing, accuracy is typically measured as the amount of similarity between an original signal and its reconstruction. This similarity is traditionally a numeric metric that does not take into account the intended purpose of the reconstructed images. In the field of medical imaging, a reconstructed image is meant to aid a physician to perform a diagnostic task. Therefore, the quality of the reconstruction should be measured by how much it helps to perform the diagnostic task. A model observer is a computer tool that aims to mimic the performance of human observer, usually a radiologist, at a relevant diagnosis task. In this work we present our linear model observer designed to automatically select the features needed to model a human observer response. This is a novelty from the model observers currently being used in the medical imaging field, which instead usually have ad-hoc chosen features. Our model observer dependents only on the resolution of the image, not the type of imaging technique used to acquire the image.
Show less

Title: Removing Confounds in Text Classification for Computational Social Science
Creator: Landeiro Dos Reis, Virgile
Date: 2018
Description: Nowadays, one can use social media and other online platforms to communicate with friends and family, write a review for a product, ask...
Show moreNowadays, one can use social media and other online platforms to communicate with friends and family, write a review for a product, ask questions about a topic of interest, or even share details of private life with the rest of the world. The ever-increasing amount of user-generated content has provided researchers with data that can offer insights on human behavior. Because of that, the field of computational social science - at the intersection of machine learning and social sciences - has soared in the past years, especially within the field of public health research. However, working with large amounts of user-generated data creates new issues. In this thesis, we propose solutions for two problems encountered in computational social science and related to confounding bias.First, because of the anonymity provided by online forums, social networks, or other blogging platforms through the common usage of usernames, it is hard to get accurate information about users such as gender, age, or ethnicity. Therefore, although collecting data on a specific topic is made easier, conducting an observational study with this type of data is not simple. Indeed, when one wishes to run a study to measure the effect of a variable on another variable, one needs to control for potential confounding variables. In the case of user-generated data, these potential confounding variables are at best noisily observed or inferred and at worst not observed at all. In this work, we wish to provide a way to use these inferred latent attributes in order to conduct an observational study while reducing the effect of confounding bias as much as possible. We first present a simple matching method in a large-scale observational study. Then, we propose a method to retrieve relevant and representative documents through adaptive query building in order to build the treatment and control groups of an observational study.Second, we focus on the problem of controlling for confounding variables when the influence of these variables on the target variable of a classification problem changes over time. Although identifying and controlling for confounding variables has been assiduously studied in empirical social science, it is often neglected in text classification. This can be understood by the fact that, if we assume that the impact of confounding variables does not change between the training and the testing data, then prediction accuracy should only be slightly affected. Yet, this assumption often does not hold when working with user-generated text. Because of this, computational science studies are at risk of reaching false conclusions when based on text classifiers that are not controlling for confounding variables. In this document, we propose to build a classifier that is robust to confounding bias shift, and we show that we can build such a classifier in different situations: when there are one or more observed confounding variables, when there is one noisily predicted confounding variable, or when the confounding variable is unknown but can be detected through topic modeling.
Show less

Title: A Complete Machine Learning Approach for Predicting Lithium-Ion Cell Combustion
Creator: Almagro Yravedra, Fernando
Date: 2020
Description: The object of the herein thesis work document is to develop a functional predictive model, able to predict the combustion of a US18650 Sony...
Show moreThe object of the herein thesis work document is to develop a functional predictive model, able to predict the combustion of a US18650 Sony Lithium-Ion cell given its current and previous states. In order to build the model, a realistic electro-thermal model of the cell under study is developed in Matlab Simulink, being used to recreate the cell's behavior under a set of real operating conditions. The data generated by the electro-thermal model is used to train a recurrent neural network, which returns the chance of future combustion of the US18650 Sony Lithium-Ion cell. Independently obtained data is used to test and validate the developed recurrent neural network using advanced metrics.
Show less

Title: IMPACT OF DATA SHAPE, FIDELITY, AND INTER-OBSERVER REPRODUCIBILITY ON CARDIAC MAGNETIC RESONANCE IMAGE PIPELINES
Creator: Obioma, Blessing Ngozi
Date: 2020
Description: Artificial Intelligence (AI) holds a great promise in the healthcare. It provides a variety of advantages with its application in clinical...
Show moreArtificial Intelligence (AI) holds a great promise in the healthcare. It provides a variety of advantages with its application in clinical diagnosis, disease prediction, and treatment, with such interests intensifying in the medical image field. AI can automate various cumbersome data processing techniques in medical imaging such as segmentation of left ventricular chambers and image-based classification of diseases. However, full clinical implementation and adaptation of emerging AI-based tools face challenges due to the inherently opaque nature of such AI algorithms based on Deep Neural Networks (DNN), for which computer-trained bias is not only difficult to detect by physician users but is also difficult to safely design in software development. In this work, we examine AI application in Cardiac Magnetic Resonance (CMR) using an automated image classification task, and thereby propose an AI quality control framework design that differentially evaluates the black-box DNN via carefully prepared input data with shape and fidelity variations to probe system responses to these variations. Two variants of the Visual Geometric Graphics with 19 neural layers (VGG19) was used for classification, with a total of 60,000 CMR images. Findings from this work provides insights on the importance of quality training data preparation and demonstrates the importance of data shape variability. It also provides gateway for computation performance optimization in training and validation time.
Show less

Title: REVEALING LINGUISTIC BIAS
Creator: Karmarkar, Sathyaveer S.
Date: 2021
Description: Readers currently face bias in articles written by writers who focus more on partiality towards any person or organization than showing the...
Show moreReaders currently face bias in articles written by writers who focus more on partiality towards any person or organization than showing the real facts. The study aims to detect and reveal such bias against them and try to portray real facts without any partiality against any person or organization. The data is fetched by selecting various articles from Google, especially those containing some bias in them. The bias was checked by measuring the subjectivity and polarity of the article using multiple libraries such as NLTK etc. We created a google form to take readers’ views showing them randomly either the biased article or the improved article after changing bias and getting their opinions.
Show less

Title: Towards Assisting Human-Human Conversations
Creator: Nanaware, Tejas Suryakant
Date: 2021
Description: The idea of the research is to understand the open-topic conversations and ways to provide assistance to humans who face difficulties in...
Show moreThe idea of the research is to understand the open-topic conversations and ways to provide assistance to humans who face difficulties in initiating conversations and overcome social anxiety so as to be able to talk and have successful conversations. By providing humans with assistive conversational support, we can augment the conversation that can be carried out. The AdvisorBot can also help to reduce the time taken to type and convey the message if the AdvisorBot is context aware and capable of providing good responses.There has been a significant research for creating conversational chatbots in open-domain conversations that have claimed to have passed the Turing Test and can converse with humans while not seeming like a bot. However, if these chatbots can converse like humans, can they provide actual assistance in human conversations? This research study observes and improves the advanced open-domain conversational chatbots that are put in practice for providing conversational assistance.While performing this thesis research, the chatbots were deployed to provide conversational assistance and a human study was performed to identify and improve the ways to tackle social anxiety by connecting strangers to perform conversations that would be aided by AdvisorBot. Through the questionnaires that the research subjects filled during their participation, and by performing linguistic analysis, the quality of the AdvisorBot can be improved so that humans can achieve better conversational skills and are able to clearly convey their message while conversing. The results were further enhanced by using transfer learning techniques and quickly improve the quality of the AdvisorBot.
Show less

Title: A Hybrid Data-Driven Simulation Framework For Integrated Energy-Air Quality (iE-AQ) Modeling at Multiple Urban Scales
Creator: Ashayeri, Mehdi
Date: 2020
Description: To date, limited work has been done to collectively incorporate two key urban challenges: climate change and air pollution for the design of...
Show moreTo date, limited work has been done to collectively incorporate two key urban challenges: climate change and air pollution for the design of sustainable and healthy built environments. Main limitations to doing so include the existence of large spatiotemporal gaps in local outdoor air pollution data and a lack of a formal theoretical framework to effectively integrate localized urban air pollution data into sustainable built environment design strategies such as natural ventilation in buildings. This work hypothesizes that emerging advanced computational modeling approaches, including artificial intelligence (AI) and machine learning (ML) techniques, along with big open data set initiatives, can be used to fill some of those gaps. This can be achieved if urban air quality explanatory factors are properly identified and effectively connected to the current building performance simulation workflows.Therefore, the primary objective of this dissertation is to develop a hybrid AI-based data-driven simulation framework for integrated Energy-Air Quality (iE-AQ) modeling to quantify the combined energy reduction profiles and health risks implications of sustainable built environment design. This framework (1) incorporates dynamic human-centered factors, including mobility and building occupancy among others into the model, (2) interlinks land use regression (LUR), inverse distance weighting (IDW), and building energy simulation (BES) approaches via the R computational platform for developing the model, and (3) develops a web-based platform and interactive tool for visualizing and communicating the results. A series of novel machine learning approaches are tested within the workflow to improve efficiency and accuracy of the simulation model. A multi-scale model of urban air quality (using PM2.5 concentrations as the end point) and weather localization model with high spatiotemporal resolution was developed for Chicago, IL using low-cost sensor data. The integrated energy and air quality model was tested for the prototype office building at multiple urban scales in Chicago through applying air pollution-adjusted natural ventilation suitable hours.Results showed that the proposed ML approaches improved model accuracy above traditional simulation and statistical modeling approaches and that incorporating dynamic building-related factors such as human activity patterns can further improve urban air quality prediction models. The results of integrated energy and air quality (iE-AQ) analysis highlight that the energy saving potentials for natural ventilation considering local ambient air pollution and micro-climate data vary from 5.2% to 17% within Chicago. The proposed framework and tool have the potential to aid architects, engineers, planners and urban health policymakers in designing sustainable cities and empowering analytical solutions for reducing human health risk.
Show less

Title: A SCALABLE SIMULATION AND MODELING FRAMEWORK FOR EVALUATION OF SOFTWARE-DEFINED NETWORKING DESIGN AND SECURITY APPLICATIONS
Creator: Yan, Jiaqi
Date: 2019
Description: The world today is densely connected by many large-scale computer networks, supporting military applications, social communications, power...
Show moreThe world today is densely connected by many large-scale computer networks, supporting military applications, social communications, power grid facilities, cloud services, and other critical infrastructures. However, a gap has grown between the complexity of the system and the increasing need for security and resilience. We believe this gap is now reaching a tipping point, resulting in a dramatic change in the way that networks and applications are architected, developed, monitored, and protected. This trend calls for a scalable and high-fidelity network testing and evaluation platform to facilitate the transformation from in-house research ideas to real-world working solutions. With this objective, we investigate means to build a scalable and high-fidelity network testbed using container-based emulation and parallel simulation; our study focuses on the emerging software-defined networking (SDN) technology. Existing evaluation platforms facilitate the adoption of the SDN architecture and applications to production systems. However, the performance of those platforms is highly dependent on the underlying physical hardware resources. Insufficient resources would lead to undesired results, such as low experimental fidelity or slow execution speed, especially with large-scale network settings. To improve the testbed fidelity, we first develop a lightweight virtual time system for Linux container and integrate the system into a widely-used SDN emulator. A key issue with an ordinary container-based emulator is that it uses the system clock across all the containers even if a container is not being scheduled to run, which leads to the issue of both performance and temporal fidelity, especially with high workloads. We investigate virtual time approaches by precisely scaling the time of interactions between containers and physical devices. Our evaluation results indicate a definite improvement in fidelity and scalability. To improve the testbed scalability, we investigate how the centralized paradigm of SDN can be utilized to reduce the simulation workload. We explore a model abstraction technique that effectively transforms the SDN network devices to one virtualized switch model. While significantly reducing the model execution time and enabling the real-time simulation capability, our abstracted model also preserves the end-to-end forwarding behavior of the original network.With enhanced fidelity and scalability, it is realistic to utilize our network testbed to perform a security evaluation of various SDN applications. We notice that the communication network generates and processes a huge amount of data. The logically-centralized SDN control plane, on the one hand, has to process both critical control traffic and potentially big data traffic, and on the other hand, enables many efficient security solutions, such as intrusion detection, mitigation, and prevention. Recently, deep neural networks achieve state-of-the-art results across a range of hard problem spaces. We study how to utilize the big data and deep learning to secure communication networks and host entities. For classifying malicious network traffic, we have performed the feasibility study of off-line deep-learning based intrusion detection by constructing the detection engine with multiple advanced deep learning models. For malware classification on individual hosts, another necessity to secure computer systems, existing machine learning-based malware classification methods rely on handcrafted features extracted from raw binary files or disassembled code. The diversity of such features created has made it hard to build generic malware classification systems that work effectively across different operational environments. To strike a balance between generality and performance, we explore new graph convolutional neural network techniques to effectively yet efficiently classify malware programs represented as their control flow graphs.
Show less

Title: Machine Learning for NDE Imaging Applications
Creator: Zhang, Xin
Date: 2023
Description: Infrared Thermography and Ultrasonic Imaging of materials are promising non-destructive evaluation (NDE) methods but signals face challenges...
Show moreInfrared Thermography and Ultrasonic Imaging of materials are promising non-destructive evaluation (NDE) methods but signals face challenges to be analyzed and characterized due to the nature of complex signal patterns and poor signal-to-noise ratios (SNR). Industries such as nuclear energy, are constructed with components produced using high-strength superalloys. These metallic components face challenges for wide deployment because material defects and mechanical conditions need to be non-destructively evaluated to identify potential danger before they enter service. Low NDE performance and lack of automation, particularly considering the complex environment in the in-situation NDE and nuclear power plant, present a major challenge to implement conventional NDE. This study solves the problems of using the advantages of machine learning as signal processing methods for Infrared Thermography and Ultrasonic NDE imaging applications. In Pulsed Infrared Thermography (PIT), for quality control of metal additive manufacturing, we proposed an intelligent PIT NDE system and developed innovative unsupervised learning models and thermal tomography 3D imaging algorithms to detect calibrated internal defects (pores) of various sizes and depths for different nuclear-grade metallic structures. Unsupervised learning aims to learn the latent principal patterns (dictionaries) in PIT data to detect defects with minimal human supervision. Difficulties to detect defects by using PIT are thermal imaging noise patterns; uneven heating of the specimen; defects of micron-level size with overly weak temperature signals and so on. The unsupervised learning methods overcome these barriers and achieve the high defect detection accuracies (F-score) of 0.96 to detect large defects and 0.89 to detect microscopic defects, and can successfully detect defects with diameter of only 0.101-mm. In addition, we researched and developed innovative unsupervised learning models to compress high-resolution PIT imaging data and achieve the average high compression ratio >30 and a highest compression of 46 with reconstruction accuracy peak signal-to-noise ratio (PSNR) >73dB while preserving weak thermal features corresponding to microscopic defects. In ultrasonic NDE imaging, for structural health monitoring of materials, we built a high-performance ultrasonic computational system to inspect the integrity of high-strength metallic materials which are used in high-temperature corrosive environments of nuclear reactors. For system automation, we have been developing neural networks with various architectures for grain size estimation by characterizing the ultrasonic backscattered signals with high accuracy and data-efficiency. In addition, we introduce a response-based teacher-student knowledge distillation training framework to train neural networks and achieve 99.27% characterization accuracy with a high image processed throughput of 192 images/second on testing. Furthermore, we introduce a reinforcement learning based neural architecture search framework to automatically model the optimal neural networks design for ultrasonic flaws detection. At last, we comprehensively researched the performance of using unsupervised learning methods to compress 3D ultrasonic data and achieve high compression performance using only 4.25% of the acquired experimental data.
Show less

repository.iit

Search the repository

Enabled Filters

Refine Results

Type

Date

Subject

Creator

Rights