21793 research outputs found
Sort by
Human Hair in Artworks by Sheela Gowda and Doris Salcedo: Engaging with Materiality and Memory
Get PDFThis thesis examines the use of human hair in sculptural installations by Indian artist Sheela Gowda (b. 1957) and Colombian artist Doris Salcedo (b. 1958), focusing on questions of materiality and memory. In Gowda’s Behold (2009), car bumpers are suspended from the ceiling by a continuous, four-kilometer hair rope made of rewoven traditional amulets, while in Salcedo’s Unland (1995-98), many individual strands of hair are stitched through the wood of disfigured tables. Despite the cultural distance between the two artists, their artworks have circulated internationally, and there is a striking commonality to how the artists explore the unusual and unique qualities of human hair as an artistic material. This thesis deploys new materialist and phenomenological methodologies, drawing on Jane Bennett’s concepts of “vibrant materiality” and “assemblage” to address Gowda’s Behold, and Sara Ahmed’s notion of “queer phenomenology” to address Salcedo’s Unland. These approaches illuminate how Gowda and Salcedo engage with their materials, and help to explain the complex impact those materials have on viewers. The thesis closes with a discussion of memory, examining the artists’ use of materials, and hair in particular, as monument-like ways of addressing collective memory.
Keywords
Contemporary art, Installation, Sculpture, Found objects, Hair, Sheela Gowda, Doris Salcedo, New materialism, Phenomenology, Memory, India, Colombia, Counter-monumen
Development of Novel Energy-Based Displacement Estimation Methods: From Ultrasound Elastography to Super-Resolution Ultrasound
Get PDFUltrasound is the second most frequently used medical imaging modality that is inexpensive, non-invasive, portable, and fast. As a real-time imaging system, ultrasound can also be used for tracking motion patterns to complement anatomical images. Temporal tracking of tissue motion, a non-trivial task, plays a pivotal role in many diagnostic applications of ultrasound, such as elastography and super-resolution ultrasound. Elastography is a non-invasive medical imaging technique that estimates tissue elastic properties to detect abnormalities in an organ. Ultrasound radio-frequency (RF) data can be tracked to compute tissue strain (which is a surrogate for elasticity) using energy-based algorithms. A continuity constraint along with the data similarity is imposed to obtain a unique solution to the displacement estimation problem. Existing energy-based methods consider only amplitude similarity to formulate the data function, which makes the displacement estimation process sensitive to outliers. In addition, they exploit the L2-norm of the first-order spatial derivative of the displacement field to construct the regularizer. This regularization scheme is not entirely consistent with the mechanics of tissue deformation while perturbed by an external force. Moreover, the L2-norm often over-penalizes the displacement discontinuity. Consequently, state-of-the-art techniques estimate noisy strain images with low target-background contrast and blurry inclusion edges. Another well known limitation of the existing displacement tracking techniques is their poor lateral estimation ability.
Ultrasound localization microscopy (ULM) is a promising medical imaging modality that systematically leverages the advantages of contrast-enhanced ultrasound (CEUS) to surpass the diffraction barrier and delineate the microvascular map. Localization and tracking of intravenously injected microbubble (MB) contrast agents, two significant steps of ULM, facilitate generating the vascular map and the velocity distribution, respectively. The existing MB tracking algorithms predominantly incorporate template-matching and bipartite graph-based cost table minimization, disregarding the immense potential of an energy-based analytic framework.
Herein, we propose six novel energy-based strain imaging techniques (Chapters 2 to 6) and one analytic optimization-based MB tracking algorithm (Chapter 7) to resolve the aforementioned issues of the existing techniques. All seven algorithms developed herein exhibit promising performance in synthetic and real experiments
A Compositional Learning Diagnoser for Discrete Event System
Get PDFIn this thesis, the design of model-based fault diagnosis systems for Discrete Event Systems (DES) is studied. The correct performance of a model-based diagnoser depends on the accuracy of the plant DES model used in its design. The behavior of this nominal model may differ from the actual plant behavior (i.e., the true model) due to various reasons such as modeling errors, modeling simplifications and coding errors. The difference between the nominal model and the true model may result in observations (i.e., sensor readings) that are unexpected by the diagnoser, resulting in "discrepancy."
In the literature, "learning diagnosers" have been proposed that in cases of discrepancy add transitions to the nominal DES model to account for the unexpected behavior. It is assumed that the nominal and true models have the same number of states, and their difference is in the transitions.
In general, every discrepancy can be explained by different sets of additional transitions, each representing a hypothesis. To narrow down the list of hypotheses, the principle of parsimony is used - giving preference to less complex hypotheses. After a set of hypotheses is generated, using future observations, this set is narrowed down. In some cases, the set may have to be expanded.
In this thesis, a new approach to learning diagnoser is introduced that takes advantage of the structure of the plant to generate hypotheses. This is meant to reduce the number of hypotheses and to generate hypotheses that are more likely to correctly explain the discrepancies. Specifically, the proposed method takes advantage of the fact that models are built incrementally from component models and their interactions. Hence, in the learning process, new transitions (to explain discrepancies) are added to the component models and their interactions (rather than to the complete flat model of the plant).
This results in a compositional approach to learning. In this thesis, a framework for the compositional approach is presented and the learning rules and the corresponding algorithms are developed. A case study from process control is used to illustrate the proposed approach
La plus haute uevre que seit : Pseudo traduction, mystification et métafiction dans le Roman de Troie de Benoît de Sainte Maure
Get PDFLe concept de la pseudo-traduction peut paraître simple de prime abord : il s’agirait d’un texte, rédigé et publié dans une langue donnée, dont le caractère original est dissimulé au profit d’une fiction le présentant comme une œuvre traduite à partir d’une langue étrangère. La simplicité apparente de cette définition est problématisée par le grand nombre de textes qui frôlent les limites de l’original et du traduit, parmi lesquels les premiers romans français, les romans d’Antiquité, représentent un cas limite d’un intérêt particulier en raison de leur rôle fondateur dans le développement de la subjectivité littéraire occidentale. Imprégnées du motif de la revendication de fidélité scrupuleuse à des sources gréco-latines souvent obscurcies, déformées ou simplement inexistantes, ces mises en roman cultivent l’art de l’expression poétique originale sous le couvert de l’autorité de la matière antique. Vulgaire tromperie digne d’une époque d’obscurantisme, ou signe d’une conception de la traduction radicalement différente de la nôtre? L’étude du roman d’Antiquité le plus reproduit du Moyen Âge, le Roman de Troie, à la lumière des études théoriques sur les supercheries littéraires modernes et postmodernes renforce la crédibilité d’une troisième hypothèse : celle qui ferait de ces prétentions de fidélité dans l’opération traduisante des manifestations de fiction autoréflexive apparentées à celles que l’on retrouve dans de nombreuses pseudo-traductions depuis le XVIe siècle
Against Exclusion: Intellectual and Developmental Disability Policy in Canada
Get PDFWithin Canada, the social inclusion of people with intellectual and developmental disabilities (IDD) is impeded by structural and social barriers borne from a legacy of systemic discrimination and policies of explicit exclusion. This exclusion is so ingrained that the realization of ‘full inclusion’ for this population significantly challenges dominant social and political norms. However, despite this challenge, for the past 25 years social inclusion has been at the forefront of the Canadian disability policy agenda. This dissertation poses two related research questions: (1) ‘how is social inclusion framed in the design and implementation of policies targeting people with IDD?’, and (2) ‘how do Canadian provinces differ in the effectiveness of social services that promote the social inclusion of people with IDD?’
These questions are addressed in three phases. The first phase involves a Critical Frame Analysis of IDD policy designs within relevant federal and provincial policy documents, identifying six distinct design frames. In the second phase, IDD policy implementation processes are assessed for how they (re)frame social inclusion. Empirical support is drawn from interviews and focus groups with policy actors, advocates, and service users. While the complex nature of implementation (re)framing in IDD services confounds cross-provincial comparison, this dissertation introduces a novel typology for comparing implementation decisions. It demonstrates that policy effectiveness need not be confined to mechanisms of top-down accountability and can be achieved through empowering implementers to adhere to professional norms or service user preferences.
This empirical analysis of social inclusion policy (re)framing develops a descriptive foundation to select and weight indicators used in the third phase: a multidimensional policy index. The Social Inclusion Services Index (SISI) comprises 10 indicators across 4 domains that capture the effectiveness of Canadian IDD policies in promoting social inclusion. The SISI offers insight into the finding that, despite widespread support for inclusive IDD policy, implementation has failed because of austere spending, stalled policy transitions and inflexible administrative structures, among other factors. Cross-provincial comparison, based on indicators reflecting priorities identified by study participants, highlights areas of emphasis to curtail policy failure in the promotion of social inclusion for Canadians with IDD
One-Pot Synthesis and Design of Chiral Carbon Dots Using Response Surface Methodology
Get PDFCarbon Dots (CDots) are a vibrant class of fluorescent nanoparticles with tunable physico-chemical and optical properties that owing to their extremely small sizes (<20nm), ease of fabrication, low cost of preparation, unique optical properties, and low cytotoxicity, they hold great promise in a myriad of applications. While their small size enables them to traverse through natural biological barriers, their surface chemical functionalities allow them to efficiently interact with their environment. In addition, CDots have been shown to retain some of the properties of their precursors, also known as active structure preservation during synthesis that has generated interest in exploring the potential applications of CDots.
In this work, we successfully synthesize and transfer the chiral properties of an amino acid precursor (L- and D-proline) to the surface of the CDots. The retention of the active sites potentially associated with the antibacterial properties of proline can be achieved by preserving the chirality of proline based CDots, leading to an increased ability to penetrate bacterial membranes. Instead of relying on an empirical one-factor-at-a-time approach to study the influential condition for this goal, we used Response Surface Methodology (RSM) to efficiently optimize the reaction conditions (temperature, time, and molar ratio) with the goal of preserving the maximum residual chiral signal of proline on the surface of CDots. We further evaluated the antibacterial response to the maximum preservation of the chiral properties in CDots versus its absence. Our observations evidence enhanced antibacterial activity for highly chiral samples confirming the correlation of proline activity with its stereochemistry
Data Stream Classification with Mondrian Forest under Memory Constraints
Get PDFSupervised learning algorithms generally assume the availability of enough memory to store
data models during the training and test phases. However, this assumption is unrealistic when data comes in the form of infinite data streams, or when learning algorithms are deployed on devices with reduced amounts of memory. In this manuscript, we investigate the use of data stream classification methods under memory constraints. Our investigation consists of three steps: a benchmark of models, an update of a model, and an optimization
of a trade-off. We evaluate data stream classification models with different criteria such
as classification performance or resource usage. The benchmark reveals that the Mondrian
forest, despite having state-of-the-art classification performance with unlimited memory, is
impacted by a low memory limit. We then adapt the online Mondrian forest classification
algorithm to work with memory constraints on data streams. In particular, we design five
out-of-memory strategies to update Mondrian trees with new data points when the memory
limit is reached. We evaluate our algorithms on a variety of real and simulated datasets, and
we conclude with recommendations on their use in different situations: the Extend Node strategy appears as the best out-of-memory strategy in all configurations. We identify that the memory-constrained brings a trade-off between the Mondrian forest size and its tree depth. We design an adjusting algorithm to optimize the forest size to the data stream and the memory limit and we evaluate this algorithm on similar datasets. All our methods are implemented in the OrpailleCC open-source library and are ready to be used on embedded systems and connected objects. Overall, the contributions significantly improve the performance of the Mondrian forest under memory constraints
Safeguarding Health and Well-Being in University Classrooms: Estimating CO2 Concentration Based on Air Permeability
Get PDFIndoor air quality (IAQ) in educational facilities is an important factor in student's well-being and academic achievement. Window opening and air infiltration are commonly used as the sole ventilation sources in current educational buildings, which can lead to unhealthy levels of indoor pollutants and energy waste. In this paper, the impact of infiltration and exfiltration on CO2 concentrations was evaluated in three studios at the architecture department in the American University in Cairo by the means of fan pressurization testing and onsite CO2 PPM concentration measurements. The monitoring protocol took place under normal operation schedule in which number of people in the space was recorded along with other parameters such as manual airing status by window opening. Airtightness was then assessed using pressurization test method which was carried out in these classrooms to assess air infiltration caused by envelope leakages. Two sets of linear regression models were developed to estimate CO2 concentration in the space: one for when windows are closed and another when they are opened. The two sets were then compared to understand the effect of airtightness on CO2 concentration. The findings indicated that ventilation cannot rely solely on air infiltration, and that particular controlled ventilation systems should be employed to optimize IAQ while avoiding excessive energy loss
Islands and Ellipses in 2D Dynamical Systems
Get PDFThree main results are presented in this thesis. The first is a proof of the existence of absolutely continuous invariant measures (ACIMs) for two dimensional maps supported on islands, which are small, disjoint regions of R2. The proof is computer-assisted and uses both numerical evidence and a combinatorial method. We give examples of weak chaos for which ACIMs exist: within islands there is chaos, but from a distance orbits are periodic.
The second main result is a geometrical proof of the asymptotic behavior of generalized tent maps with memory which we call elliptical maps. It is proved that for certain π-rational angles, all points in the domain except for (0, 0) fall into a polygonal region whose characteristics we determine. When the angles are π-irrational we prove that these points either fall in a unique ellipse, or accumulate on its boundary.
The third result is a proof that ACIMs exist for a certain range of parameters in generalized β-tent maps with memory.
The thesis begins with discussions about ACIMs and why they are interesting, followed by Tsujii’s theorem and other tools, notes on computer calculations and graphics, and on weak chaos. At the end, we highlight some unanswered questions and puzzling phenomena that we encountered during our research
Mixture-Based Clustering and Hidden Markov Models for Energy Management and Human Activity Recognition: Novel Approaches and Explainable Applications
Get PDFIn recent times, the rapid growth of data in various fields of life has created an immense need for powerful tools to extract useful information from data. This has motivated researchers to explore and devise new ideas and methods in the field of machine learning. Mixture models have gained substantial attention due to their ability to handle high-dimensional data efficiently and effectively. However, when adopting mixture models in such spaces, four crucial issues must be addressed, including the selection of probability density functions, estimation of mixture parameters, automatic determination of the number of components, identification of features that best discriminate the different components, and taking into account the temporal information. The primary objective of this thesis is to propose a unified model that addresses these interrelated problems. Moreover, this thesis proposes a novel approach that incorporates explainability.
This thesis presents innovative mixture-based modelling approaches tailored for diverse applications, such as household energy consumption characterization, energy demand management, fault detection and diagnosis and human activity recognition. The primary contributions of this thesis encompass the following aspects:
Initially, we propose an unsupervised feature selection approach embedded within a finite bounded asymmetric generalized Gaussian mixture model. This model is adept at handling synthetic and real-life smart meter data, utilizing three distinct feature extraction methods. By employing the expectation-maximization algorithm in conjunction with the minimum message length criterion, we are able to concurrently estimate the model parameters, perform model selection, and execute feature selection. This unified optimization process facilitates the identification of household electricity consumption profiles along with the optimal subset of attributes defining each profile. Furthermore, we investigate the impact of household characteristics on electricity usage patterns to pinpoint households that are ideal candidates for demand reduction initiatives.
Subsequently, we introduce a semi-supervised learning approach for the mixture of mixtures of bounded asymmetric generalized Gaussian and uniform distributions. The integration of the uniform distribution within the inner mixture bolsters the model's resilience to outliers. In the unsupervised learning approach, the minimum message length criterion is utilized to ascertain the optimal number of mixture components. The proposed models are validated through a range of applications, including chiller fault detection and diagnosis, occupancy estimation, and energy consumption characterization. Additionally, we incorporate explainability into our models and establish a moderate trade-off between prediction accuracy and interpretability.
Finally, we devise four novel models for human activity recognition (HAR): bounded asymmetric generalized Gaussian mixture-based hidden Markov model with feature selection~(BAGGM-FSHMM), bounded asymmetric generalized Gaussian mixture-based hidden Markov model~(BAGGM-HMM), asymmetric generalized Gaussian mixture-based hidden Markov model with feature selection~(AGGM-FSHMM), and asymmetric generalized Gaussian mixture-based hidden Markov model~(AGGM-HMM). We develop an innovative method for simultaneous estimation of feature saliencies and model parameters in BAGGM-FSHMM and AGGM-FSHMM while integrating the bounded support asymmetric generalized Gaussian distribution~(BAGGD), the asymmetric generalized Gaussian distribution~(AGGD) in the BAGGM-HMM and AGGM-HMM respectively. The aforementioned proposed models are validated using video-based and sensor-based HAR applications, showcasing their superiority over several mixture-based hidden Markov models~(HMMs) across various performance metrics. We demonstrate that the independent incorporation of feature selection and bounded support distribution in a HAR system yields benefits; Simultaneously, combining both concepts results in the most effective model among the proposed models