5063 research outputs found
Sort by
Decentralized graph-based multi-agent reinforcement learning for traffic signal optimization
No full textSignalized intersections are persistent bottlenecks where inefficient operations contribute to congestion, delays, safety risks, and environmental impacts. Conventional control strategies provide stability under predictable demand but lack the adaptability required to manage stochastic and heterogeneous traffic conditions. This dissertation develops a decentralized graph-based multi-agent reinforcement learning (DGMARL) framework for adaptive traffic signal control. The framework advances the state of the art by (i) embedding operational constraints, including minimum/maximum green durations, pedestrian recalls, and clearance intervals, directly into the learning process; (ii) modeling intersections as decentralized agents that exchange direction-specific states through multi-head graph attention to capture asymmetric flows and upstream inflows, thereby enabling scalable coordination across large networks; and (iii) incorporating contextual pedestrian demand via point-of-interest weighting. Control policies are optimized within a constrained Markov decision process, where modular phase selection and fairness-aware rewards jointly balance vehicle efficiency and pedestrian accessibility. The framework is validated using a high-fidelity digital twin–based simulator with real-world traffic data and further demonstrated through preliminary on-street field testing on the MLK Smart Corridor. In simulation, the proposed approach reduced pedestrian waiting times by up to 24.7% and vehicle delays by 22.6%, while decreasing emissions (CO, CO_2, NO_x, and PM_10) by an average of 9.6% and increasing vehicle throughput by more than 22%. These improvements were achieved while ensuring compliance with safety-critical signal timing rules. Analysis of graph attention weights highlights interpretable coordination across intersections, confirming the robustness and scalability of the decentralized design under varied traffic conditions. In field operation, the decentralized agents correctly interpreted real-time traffic demand, switched signal phases adaptively, and responded to pedestrian push-button activations within approximately 10-15 s along with proper recall logic, maintaining Safety and Priority of Timing (SPaT) compliance and end-to-end processing latency between 80 and 120 ms. Together, these contributions establish a pathway toward deployment-ready, equitable, and sustainable traffic signal control across diverse network settings
Is Teamwork Influenced by the U.S. Subculture?
No full textWork teams can be defined as groups that are “composed of individuals who (a) exist to perform organizationally relevant tasks, (b) share one or more common goals, (c) interact socially, (d) exhibit task interdependencies, (e) maintain and manage boundaries, and (f) are embedded in organizational contexts that set boundaries, constrains the tram, and influences exchanges with other units in the broader entity” according to Kozlowski and Bell (200, p. 334). Teams are a dynamic collection in the workplace that are able to undergo developmental changes and reformations in their duration of existence within the workplace. Tuckman’s (1965) Model of Group Development can be seen in “Chapter 12: Team Dynamics and Processes Within Organizations” of Organizational Psychology: A Scientist-Practitioner Approach (3rd ed.). This provides a valuable model due to the important stages of forming, storming, norming, performing, and adjourning along with the assumption that this process is a continuation and not a consistent and linear process (Jex & Britt, 2014). According to Eby & Dobbins (1997), high levels of collectivism were associated with prior success when working in teams. In a previously reported survey, 82% of United States organizations reported that their employees to some capacity belonged to a team in their work setting (Gordon, 1992). While the United States typically is considered to be an individualist culture, there is a growing section of the literature that supports the ideology that subcultures within the U.S. particularly within Appalachia align with collectivist values more than those who live outside of the region (Gore & Wilburn, 2010). Previous literature distinguishes between the larger United States and the counties that fall within the boundaries of the Appalachian Region. The Appalachian Regional Commission (ARC) is frequently referenced in research identifying key differences between Appalachian and non-Appalachian populations in an attempt to identify cultural differences, healthcare inequity and accessibility, and economic disparities throughout the mountain range as it differs from the rest of the United States (Seufert & Carrozza, 2004). The labor market in the Appalachian region yields a complex history of blue-collar occupations and the coal mining industry having a large impact on the Appalachian workforce (Seufert & Carrozza, 2004). There is a focus in the literature on the impact that the coal mining industry had on poverty, employment, and health within this distinct population (Lobao et al., 2016). Within Appalachia, there are three subregions which can be identified as northern, central, and southern regions (Appalachian Region Commission, n.d.). Previous studies have used different subregions of Appalachia to draw a comparison of Appalachian culture to hold collectivist values more so than the larger populations within the United States (Gore & Wilburn, 2010). When considering the similarities in values between collectivist values and Appalachian values, it is then reasonable to consider if this alignment is an influence on the effectiveness of teamwork within the Appalachian workplaces divided into the three subsections of the Appalachian region. Because it has been demonstrated that teamwork is prevalent amongst American workplaces, there is a contrast to be drawn between non-Appalachian and Appalachian workers to consider the efficiency of Appalachian teamwork in the workplace environment influenced by collectivist values
Unified Mathematical Modeling, Analysis and Simulation of COVID-19 and Ecological Processes
Full text linkThis dissertation is concerned with the modeling, simulation, and analysis of complex population dynamics using compartmental modeling frameworks based on ordinary differential equations (ODEs). First, we propose a mechanistic model to investigate the transmission dynamics of COVID-19, particularly focusing on the emergence and population-level impact of a post-acute sequela referred to as long COVID and vaccination effects. We then move forward with a model that incorporates multiple strains and reinfection dynamics. By fitting these models to epidemiological data from the United States and the United Kingdom, we conduct both mathematical analyses and numerical simulations to better understand the factors driving long COVID prevalence and the inter-strain dynamics of COVID-19 that affect transmission. In addition to epidemiological applications, we extend our compartmental modeling approach to ecological studies, specifically examining the group dynamics of degus, small rodents with complex social behaviors. We analyze how environmental factors such as temperature variations, seasonal rainfall patterns, and vegetation indices (NDVI) influence group sizes and social interactions over time. This ecological modeling employs similar mechanistic ODE-based frameworks, highlighting methodological consistency across biological systems
A Combination of Modified Stacking Method with Voting Ensemble Technique for Binary Classification
Full text linkBinary classification is a frequently encountered machine learning problems. Many machine learning algorithms are used to solve binary classification problems. However, all machine learning algorithms are not suited or efficient for solving binary classification problems.
In this paper, authors has proposed a approach. The proposed approach is a machine learning algorithm, to solve binary classification problems. Authors has combined modified stacking and voting ensemble machine learning techniques for binary classification problems. Before combining the ensemble techniques, authors has modified the stacking ensemble technique for making the proposed method more robust.
This proposed method was applied on two different binary class data set to generalize. The experimental result of the proposed method shows that it can effectively classify binary class data set with an improved accuracy of 99.63%.
Moreover, it shows better performance than other state-of-the-art machine learning algorithms for binary classification
Modeling the dynamics of user adoption and abandonment for a single product
Full text linkWe introduce a compartmental differential equation model to study the dynamics of user adoption and abandonment for a single product. The model integrates two forms of abandonment: infectious, driven by user interactions, and non-infectious, prompted by external influences. Notably, the infectious abandonment coefficient varies linearly with the number of previous users. We investigate the existence of equilibria of the model and derive the threshold quantity ℛ0. The user-free equilibrium is always present, and its stability is analyzed under the condition ℛ0 \u3c 1. Moreover, a user-prevailing equilibrium does not exist when ℛ0 ≤ 1, but at least one user-prevailing equilibrium is guaranteed when ℛ0 \u3e 1. We further characterize conditions for multiple equilibria and various bifurcations, including saddle-node, -shaped, and Hopf bifurcations, and formulate an optimal control problem. Numerical simulations validate our theoretical findings, and the historical LinkedIn and YouTube data calibrate the model to forecast future user adoption trends
The time-independent Schrodinger equation with Dirac delta potentials in an infinite potential well
Full text linkThe infinite potential well is one of the most well-known models in quantum mechanics, as well as the Dirac delta potential. In the first part of the thesis, one Dirac delta potential is considered using the time-independent Schr\ {o}dinger equation, the typical boundary conditions for an infinite well and normalization conditions to obtain the wave function . The system naturally imposes an additional condition at due to the Dirac delta function, which yields the final equation for the energy levels that has infinitely many solutions. In the second part, a lattice potential within the infinite well is considered, which is defined as a finite sum of Dirac delta functions that are spread evenly within the bounds of the well. For future work, more numerical analysis should be done on these systems, as well as expanding the problem to several dimensions
Public memory of World War II in Japan
Full text linkPublic memory often fails to encompass varied perspectives on past events, representing exclusively the collective understanding of one group or nation. Most nations shape history through personalized narratives, which remember their own motivations in conflicts, neglect their harms done against others, or memorialize past harms and traumas for reasons such as collective identity, nationalism, or guilt. However, while public memory is similarly controlled from nation to nation, various scholarship has illuminated changes within Japanese World War II education, and many scholars have discussed its potential impact to Japanese World War II public memory. To fully analyze Japanese public memory and the influence of these changes, this thesis compiled interviews from Japanese college students concerning their public memory of World War II. Through these interviews, this paper examined current Japanese public memory of World War II and its influencing factors, specifically Japanese education, museums and memorials, and popular culture. This thesis identifies three distinct historical narratives of justification, victimization, and pacifism throughout present-day Japanese public memory and argues nationalists utilize these historical narratives to shape public memory and unify Japanese national identity
Technology and transportation’s role in social isolation in rural older adults
Full text linkThis study examines the perceptions and experiences of rural older adults. More specifically, their community experiences, connections, technology, and transportation. Past literature, when examining these topics, took a quantitative approach utilizing scales. This study takes a qualitative approach, utilizing semi-structured interviews to examine the feelings behind the experiences of the participants. Interviews were conducted with four older rural adults. The four participants were female and were of varying marital statuses. Three of the participants were Caucasian, and one was African American. Four themes were found in this process. It was found that the participants were content with the connections that they had, alluding to the quality of connections outweighing the quantity of connections. The theme of community awareness was evident in how the participants were diligent in connecting to their communities. The fear and education of technology were found as themes in the interviews. It was found that older adults were scared of messing up technology while simultaneously not wanting education on technology. The last theme was concerns of public transportation, meaning that there were needs for public transportation and improvements that needed to be made to existing infrastructure. The themes point to the need to simplify practices when utilizing technology and increase communication regarding public transit
Unleashing movement: analyzing predictors of children’s physical activity in a children’s museum environment
Full text linkThis study examines the role of an interactive children\u27s museum in promoting physical activity among children aged 5-10, with a focus on the influence of various predictors, including sleep, family activity, typical physical activity levels, and participation in sports. Using a mixed-method design, participants were monitored with an Actigraph wGT3X accelerometer during a typical museum visit, and their parents completed a survey on factors influencing their child\u27s activity. The study had 15 participants who were recruited by a member of the research team as they entered the museum. The results revealed that children whose parents were more active with them exhibited higher levels (14% of their visit spent in moderate or vigorous activity) of physical activity during the museum visit. Additionally, children who typically engaged in more than 60 minutes of physical activity per day showed greater physical activity (14% of time spent in moderate or vigorous activity) in the museum. Unexpectedly, factors like sleep, school type, and participation in sports did not significantly impact physical activity levels in this setting. This study underscores the potential of children’s museums as informal learning spaces that can enhance physical activity, highlighting the importance of family involvement and habitual physical activity. These findings can inform future interventions to promote healthier lifestyles for children
Christianity and environmental stewardship: locating a new Christian environmental ethic in response to the climate crisis
Full text linkIn Lynn White Jr.’s essay, “The Historical Roots of Our Ecological Crisis,” White argues that Christianity has heavily influenced society’s view of the environment. Rhetoric from Genesis has led people to believe that God intends, and even commands, that humans have dominion over the earth. This narrative provides justification for the abuse of the environment at large, abuse that has been compacted over time to create the climate crisis that we are currently experiencing. Though White’s argument lacks some nuance and somewhat exaggerates Christianity’s role in the environmental crisis, I believe his thesis to be true: Christianity does have a negative impact on individual’s environmental attitudes. Ascribing to Christian doctrine, however, is not the sole predictor in determining one’s environmental attitude. Factors such as denominational affiliation, political beliefs, and biblical literalism also impact one’s opinions on man’s role in the environment. However, Christianity can also be used as a tool to help fight the ideas that have led to the climate crisis. Instead of focusing on passages that stress human dominion over the earth, we can look to other biblical ideas about environmental stewardship, creation care, and neighborly love to encourage a Christian environmental ethic centered around fighting the climate crisis as an act of worship. Christians of the past and present, namely St. Francis of Assisi and the Sisters of the Community of St. Mary, practice an environmental ethic centered in their religious beliefs, and their doctrine can be used as a pattern to build a modern Christian environmental ethic that is conducive to addressing the current environmental crisis