University of Illinois Urbana-Champaign
Illinois Digital Environment for Access to Learning and Scholarship RepositoryNot a member yet
123813 research outputs found
Sort by
Multi-location networks and ecosystem detection: methods and applications
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Siddhanth Sabharwal, accepted the attached license on 2025-07-02 at 11:36.The student, Siddhanth Sabharwal, submitted this Dissertation for approval on 2025-07-02 at 11:45.This Dissertation was approved for publication on 2025-07-07 at 11:45.DSpace SAF Submission Ingestion Package generated from Vireo submission #22400 on 2025-10-25 at 15:52:57We introduce multi-location networks to model the real-world phenomena of nodes existing in multiple places simultaneously. A multi-location network contains a set of locations. Each location can have a different set of nodes, but nodes can be present in multiple locations at once. In contrast, a multiplex network contains a set of layers, where each layer shares the same set of nodes. By combining these two structures, a multi-location multiplex network (MLMN) enables nodes to appear in various locations and connect through multiple edge types. For example, in a species interaction MLMN, locations could correspond to continents, while layers represent different types of interactions, such as predation and competition. Some species may inhabit multiple continents, with interactions varying by location. Analyzing the evolution of MLMNs poses substantial challenges due to the complexity of integrating location and edge type. Traditional dynamic network logistic regression models often aggregate data across these structures, leading to information loss. In this dissertation, we extend dynamic network logistic regression to accommodate MLMNs, improving the estimation of factors driving changes in such networks. Our approach predicts node presence in each location and the formation of edges across different layers at each location. We provide theoretical guarantees for the model, including global optimality, consistency, asymptotic normality, and asymptotic efficiency of the estimated parameters. The model's finite sample performance is evaluated through simulations. Additionally, a case study is conducted using a dataset from Equinix, a 2024 Fortune 500 company and one of the world's largest data center operators. Our findings highlight the strong impact of location on node presence and link formation, showcasing the practical relevance of understanding multi-location structures to capture complex network dynamics. Dynamic Generalized Multi-Location Multiplex Networks (DGMLMNs) describe a sophisticated class of networks characterized by temporal node dynamics, directed weighted edges, and concurrent multiple networks indexed by location, with nodes allowed to be in multiple locations at once. This research addresses a gap in network modeling literature by developing methods to describe these complex network structures. We propose a comprehensive approach to infer key drivers of node presence and link formation, enabling prediction of the future state of a network. The methodology employs a two-step network structure model: first modeling node presence across locations, second modeling edge counts from one node to another for each location-layer combination. Through theoretical analysis, we establish the asymptotic properties of the estimators and through simulations, we validate the finite-sample estimation and prediction performance. The model's practical utility is demonstrated through application on an anonymized DGMLMN dataset from Equinix, a 2024 Fortune 500 company, where the findings allowed for the testing and validation of various business hypotheses. As the development of methods for complex networks continues to expand, our contribution aims to open up new applications of network science for practitioners. Ecosystems, like communities, are groupings of nodes within networks. However, unlike communities, ecosystems are defined not by dense connectivity but by the nature of competition and cooperation. Drawing on ideas from management science literature, we introduce a rigorous network-based framework for defining key ecosystem concepts such as focal nodes, suppliers, customers, and complementors. These concepts give further meaning to terms like value, co-opetition, and ecosystem formation. We propose a new method, ecosystem detection, which takes a directed, weighted network as input and outputs an interpretable partition of the nodes. The method proceeds in three steps: identifying focal nodes, clustering them into ecosystems, and assigning the remaining nodes to an ecosystem. A hypothesis testing procedure to verify the presence of ecosystems within a network is also presented. We apply this method to two real-world networks. The first is the United States (U.S.) airport network, which records the count of passenger flights between airports. The second is an anonymized interconnection network provided by Equinix, a global data center operator and 2024 Fortune 500 company. In both cases, ecosystem detection yields contextually coherent partitions, offering insights beyond those provided by community detection
Toward a theory of value co-creation in digital leisure experiences
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Xin Du, accepted the attached license on 2025-07-02 at 15:07.The student, Xin Du, submitted this Dissertation for approval on 2025-07-02 at 15:07.This Dissertation was approved for publication on 2025-07-14 at 11:59.DSpace SAF Submission Ingestion Package generated from Vireo submission #22401 on 2025-10-25 at 15:52:57Value co-creation has been widely recognized and applied as a key strategy for designing personalized and memorable leisure experiences. In particular, digital platforms, along with their technological affordances, have unprecedentedly empowered leisure consumers with increasing agency to engage in and co-create their own leisure experiences. However, limited research has examined how digital service providers can effectively facilitate co-creation to enhance consumers’ digital leisure experiences. This dissertation aimed to investigate the dynamics and multi-layered nature of value co-creation in digital leisure contexts. This dissertation used an abductive approach to develop a theory elucidating the relationships between specific antecedents, interactional mechanisms, and value outcomes associated with co-creation in digital leisure. Using TikTok as the focal platform, the study explored three facets of value co-creation: (1) the technological, operational, and institutional affordances provided by the digital platform; (2) digital platform users’ engagement with platform affordances through three co-creation practices spanning consumption, creation and sharing, as well social interaction encounters; and (3) the conditions under which value co-creation failed and the recovery strategies employed by both users and the platform to restore lost values. Adopting a multi-stakeholder perspective, this dissertation drew on in-depth interviews with TikTok employees, travel content consumers, and travel content creators. Through integrating backstage platform design considerations with frontstage consumer experiences, the study offered a more holistic and practical understanding of value co-creation in digital leisure. Accordingly, the findings developed actionable insights for the design of personalized, engaging, and user-centered digital leisure experiences
De(s)colonial assemblages • life drives
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Ana Valderrama, accepted the attached license on 2025-07-12 at 17:46.The student, Ana Valderrama, submitted this Dissertation for approval on 2025-07-12 at 20:05.This Dissertation was approved for publication on 2025-07-16 at 10:26.DSpace SAF Submission Ingestion Package generated from Vireo submission #22533 on 2025-10-25 at 15:53:51My dissertation explores correlations between de(s)colonial body-artefact assemblages (Deleuze and Guattari 1980/2020) and socio-political dynamics in Latin American peripheries. I take the periphery as a personal, theoretical, and geopolitical positionality; a conceptual ethical-political category; and a physical territory shaped by a growing tension between, on the one hand, the hegemonic, multidimensional, dynamic, and fluid apparatus of the colonial-modernity project (Segato 2013/2021) and on the other hand, the ecologies of resistance, dissidence, and insurgency. Through the study of representative cases, I show how power dynamics specific to those peripheries are catalyzed and take form within representative assemblages that incarnate in their material and symbolic consistency, latent historical and political processes that are otherwise invisible. I use the term assemblage to refer to a multiplicity in which ideas, bodies (human and non-human), natural phenomena, materials, energies, and agencies with differential degrees of codification and territorialization encounter, entangle, interact, and transform each other with different intensities and velocities. I argue that the colonial-modernity project is a multidimensional, dynamic, and fluid machinery operating through four components—capitalism, racism, heteropatriarchy, and pollution—and over three targets—bodies, lands, and subjectivities. The de(s)colonial assemblages included in this study emerge from the colonial-modernity project’s cracks, fragments, remnants, and ruins. However, their agency is not fenced or entirely defined by the rules of the colonial-modernity project. As African-American Studies scholar Saidiya Hartman (1997, 56) points out, “strategies of domination don’t exhaust all possibilities of intervention, resistance, escape, refusal, or transformation.” The de(s)colonial assemblages explored in depth in my dissertation present alternatives to hegemonic discourses and hierarchical powers through their active participation in the correlation of forces, helping to unearth, visualize, and germinate critical alternatives and adding new conceptual layers to ongoing counter-hegemonic practices in the Global South. The assemblages selected for study in this dissertation have emerged from participatory, dialogic, and collective methods and actions undertaken by groups of designers and artists in collaboration with grassroots, religious, and/or political organizations, as well as members of communities in need. The processes of conception, production, expression, and reception of the selected assemblages beyond their construction have contributed to reinforcing social bonds within disregarded communities and to strengthening grassroots and social organizations. These assemblages can be considered de(s)colonial since they have been conceived as in-situ, in-vivo collective practices that reconnect bodies, lands, and subjectivities intentionally dismembered by the colonial-modernity project. Those practices are aimed at thwarting the colonial-modernity project, but they do not have enough capacity to accomplish that in one fell swoop, nor do they constitute definitive, heroic, or performative fulfillments of structural struggles. Thus, they cannot be conceptualized as revolutionary in the orthodox way. Yet they constitute a constellation of quotidian manifestations of life that disturb the trajectory of the colonial-modernity’s devices. We may conceive them as noises, holes, pricks, or as experiments in perpetual inconclusiveness. The practices in question negotiate, deflect, distort, twist, and even cannibalize the devices of the colonial-modernity project itself while triggering new scenarios within a swarming correlation of forces. Given their vitality and ingenuity, I conceive them as life drives: ecologies of resistance, dissidence, and insurgency. Understanding the complexity of these assemblages requires both epistemological and ontological reflection, focusing at once on their and my own positionality and political and cultural directionality and their being itself—that is, their processes of conception, expression, and reception. The research strategy for this work consists of constructing a relational map of de(s)colonial assemblages in which the correlation of forces in the historical present of Latin American peripheries resonates. To that end, I have applied a maximum variation sampling strategy (Patton Quinn 1987) in which I have selected eleven representative cases: Colectivo Pico in Venezuela; Escuela Radical and Taller Experimental in Mexico; Arquitetura na Periferia, A Rocinha Resiste, and Grupo Eco in Brazil; Arquitectura Expandida in Colombia; and Arquitectura del Sur, Matéricos Periféricos, CAPA, and A77 in Argentina. The cases have been selected as they represent well a particular form of association and specific forms of resistance, dissidence, or insurgency relative to symbolic or material components of the colonial-modernity project: capitalism, racism, heteropatriarchy, and pollution. Among the many assemblages produced by these groups, I have selected a total of twenty-two for this study. From a methodological viewpoint, these assemblages can be understood as philosopher and semiotician Roland Barthes’ punctums (1980/2010) since they constitute little rebel details that disturb the apparent uniformity and coherence of the general picture while exposing latent historical and political processes that are otherwise invisible. To study the selected assemblages, I have applied a situated qualitative method involving a succession of distinct iterative stages that inform one another. Data have been collected through direct observation of the sample cases in the form of bodily immersion and in-situ, in-vivo engagement guided by scores, in-person structured and unstructured interviews, and the recording and transfer of sensory experiences in various media. Data have been collected under IRB protocols, systematized, and intersected with literature review and consultation of historical archival materials. The research findings described above are then synthesized and disclosed in a non-linear set of narratives (Ch. 4: Findings. Constellation of Parables). Critical reflections on the implications of the findings are offered in chapter 5 (Ch. 5: Implications. Glossary of Life Drives), and further discussions are provided in the form of scores (Ch. 6: Discussion. Scores for reconnecting bodies, lands, and subjectivities) for people to perform in similar situations. The contents of each section are presented in a non-linear way. Readers are encouraged to explore and discern constellations of meaning from their individual and shared perspectives. The aim of this approach is to leave the conclusion of the dissertation open, blurring any attempt to achieve a totality or a unique truth. Through this dissertation, I expect to contribute to the debate on de(s)colonial thought in Latin America and the Global South. Amplifying the voices emerging from the de(s)colonial assemblages included in my study is a key objective of my dissertation. Critical reflection on those “life drives” and their agency, imagination, goals, and capacity for organization may help re-connect vital bonds among bodies, lands, and subjectivities. Also, this dissertation can help build and support reciprocal and solidarity networks among communities, organizations, scholars, and activists. Nurturing intercultural coalitions in which individuals and groups from different ontological origins and identities get “together (yet) distinct” (Chadwick 2012, xiii) may help trigger future emancipatory projects in defense of life in all its forms
Experimental investigation of two-phase immersion cooling of cylindrical lithium-ion cell with a low boiling point dielectric fluid
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Harsh Tyagi, accepted the attached license on 2025-07-20 at 18:08.The student, Harsh Tyagi, submitted this Thesis for approval on 2025-07-20 at 18:16.This Thesis was approved for publication on 2025-07-21 at 16:38.DSpace SAF Submission Ingestion Package generated from Vireo submission #22668 on 2025-10-25 at 15:54:23Electric vehicles (EVs) are at the forefront of sustainable transportation, requiring advancements in battery technology to enable faster charging speeds, longer lifespan, and enhanced safety. As temperature plays a critical role in the efficiency and longevity of lithium-ion cells, a key challenge to EV battery performance is effective thermal management. Elevated temperatures accelerate internal reactions, leading to faster aging, while low temperatures increase internal resistance and reduce capacity due to decreased ion mobility. With EVs operating in extreme climates and under increasing power demands, thermal management technologies must evolve to meet the growing cooling and heating requirements. This thesis explores the application of 2-phase immersion cooling as a solution to address the thermal challenges faced by EV battery packs. The study begins with a thorough review of existing research on 2-phase immersion cooling, followed by the design of an experimental setup to study 2-phase immersion cooling of a cylindrical lithium-ion cell. Experimental investigations assess the thermal performance of this cooling technique under high C-rates, with performance comparisons made between 0%, 50%, and 100% immersion of the cell. The results demonstrate significant improvements in cooling performance, with the 100% immersion case maintaining the mean cell temperature under 31.4 °C even during an aggressive 10C discharge, compared to 39.6 °C under the 50% immersion condition. However, for the 0% immersion case, the mean cell temperature exceeds the safe operational limit, reaching 88.4 °C during an 8C discharge. Dynamic testing further shows enhanced temperature stability under time-varying electric loads with 100% immersion, in contrast to the 0% immersion scenario. This research highlights the potential of 2-phase immersion cooling for improving the thermal performance of EV battery systems. Future work is outlined, focusing on cyclic ageing and battery pack level performance for this promising technology
Intelligent packaging: time-temperature indicators for monitoring the quality of pasteurized milk
No full textSubmission published under a 24 month embargo labeled 'Closed Access', the embargo will last until 2027-08-01The student, Lin Li, accepted the attached license on 2025-07-23 at 17:09.The student, Lin Li, submitted this Thesis for approval on 2025-07-23 at 17:16.This Thesis was approved for publication on 2025-07-25 at 10:09.DSpace SAF Submission Ingestion Package generated from Vireo submission #22715 on 2025-10-25 at 15:54:35Pasteurized milk is a highly nutritious but perishable product whose quality can rapidly deteriorate under improper storage conditions. Temperature fluctuations during storage and transportation often go unnoticed. Therefore, real-time monitoring of temperature exposure is essential to ensure food quality. Time–temperature indicators (TTIs) offer a promising, non-destructive approach for monitoring the combined effect of time and temperature on food products by recording their temperature history throughout storage and distribution. By providing a real-time signal of temperature-induced quality loss, TTIs help assess product quality throughout storage and distribution. This technology can effectively reflect changes in food quality, showing great potential to reduce food waste and improve consumer confidence. This study investigates the development of a TTI system for pasteurized milk, utilizing nanomaterials and a biopolymer to generate non-reversible colorimetric responses that reflect cumulative time and temperature exposure. Unlike conventional date-labeling systems, which do not account for real-time temperature abuse during storage or transport, TTIs provide a dynamic and visual means of monitoring food quality. The proposed system aims to serve as a non-destructive and user-friendly tool to inform consumers of potential quality degradation
Asymptotic Convexity of Wide and Shallow Neural Networks
Get PDFThere has been considerable interest in analyzing the observed empirical success of wide neural networks, both shallow and deep. A small sample of the enormous activity in this domain can be found in [2], [3], [5]–[8], [10], [14], [15]. In this short note, for a simple model of shallow and wide networks, we establish asymptotic convexity for the maps of error as a function of the parameters, equivalently their truncated epigraphs (defined below) using a known convexification effect of Minkowski sums of compact sets in Rd. In particular, this suggests that the limiting optimization problem implicit in neural network training for the infinitely wide neural network is a simple minimization problem, therefore amenable to SGD: all local minima are global minima. While the latter property is a consequence of convexity, it does not imply convexity and it may begin to hold for neural networks with ‘sufficiently large width’. This leads to a plausible explanation of the empirically observed good performance of shallow and wide networks
Linking Diffusion Models & I-MMSE
Get PDFThis paper establishes a theoretical connection between diffusion models and the I-MMSE theorem in information theory. Specifically, we demonstrate that two widely used log-likelihood-based training objectives in diffusion models are closely related, with their relationship explicitly established via the I-MMSE theorem. As a direct consequence of this theoretical link, we derive a single-step denoising method that achieves optimal performance without iterative sampling
Communication Systems with Partially Existing Encoder and Decoder
Get PDFDesign of the optimal encoder and decoder for a discrete memoryless channel is a well-studied problem in classic information theory. However, there exists important systems which operate with components (encoder and decoders) that are prebuilt and possibly mismatched, limiting the achievable performance metrics. In this paper, we propose a model that takes into account such constraints. Using this model, we develop an understanding of the source coding problem in this context and introduce two new functions -transitory and restricted rate distortion- and analyze certain properties of them. In addition, one important application of this model for developing a comprehensive theory of semantic communication is discussed. Finally, some numerical examples are provided to illustrate the concepts mentioned in the paper
The H-property of Asymmetric Step-graphons
Get PDFWe sample directed graphs from asymmetric step-graphons and investigate the probability that the random digraph has at least one node-wise Hamiltonian decomposition. We show that for almost all step-graphons, the probability converges to either zero or one as the order of the digraph goes to infinity. We exhibit a set of conditions that can essentially decide the zero-one law, and numerically validate the result
Optimizing Preventive and Reactive Defense Resource Allocation with Uncertain Sensor Signals
Get PDFCyber attacks continue to be a cause of concern despite advances in cyber defense techniques. Although cyber attacks cannot be fully prevented, standard decision-making frameworks typically focus on how to prevent them from succeeding, without considering the cost of cleaning up the damages incurred by successful attacks. This motivates us to investigate a new resource allocation problem formulated in this paper: The defender must decide how to split its investment between preventive defenses, which aim to harden nodes from attacks, and reactive defenses, which aim to quickly clean up the compromised nodes. This encounters a challenge imposed by the uncertainty associated with the observation, or sensor signal, whether a node is truly compromised or not; this uncertainty is real because attack detectors are not perfect. We investigate how the quality of sensor signals impacts the defender’s strategic investment in the two types of defense, and ultimately the level of security that can be achieved. In particular, we show that the optimal investment in preventive resources increases, and thus reactive resource investment decreases, with higher sensor quality. We also show that the defender’s performance improvement, relative to a baseline of no sensors employed, is maximal when the attacker can only achieve low attack success probabilities