Millennials are the newest generation to enter the workforce. When Millennials enter organizations, managers construct perceptions about Millennials’ communication behaviors, including their characteristics and adherence to organizational rules. These perceptions help managers decide Millennials’ organizational fit. A review of literature revealed a scarcity of empirical research in this area with little empirical research from communication scholars who apply communication frameworks, theories, and concepts. This research used the lens of social constructionism to understand the membership categorization devices and category-bound activities managers use to characterize Millennials. In order to better understand how Millennials conform to and change organizational culture, data were reviewed for those normative and code rules managers described Millennials violating. In this qualitative, exploratory study, 25 managers who were 31 years of age or older that worked in the hospitality industry and managed Millennial (18 to 30 years old) employees were interviewed through a snowball convenience sample. Interviews were transcribed and patterns were identified. Data analysis indicated that “kids,” “age group,” and “Millennials” and variations of the Millennial term were used to categorize Millennials. Analysis of category-bound activities showed patterns in Millennials’ desire for learning and training, mixed preference for teamwork often affected by their liking for peers, and needs for frequent, clear, personalized feedback. With respect to rule violations, data showed that some organizations were adapting their cell phone policies in response to Millennial rules resistance. However, organizations were not willing to accommodate Millennials’ rule violations in either the area of time-off requests or uncivil behavior due to organizational codes. iv Keywords: Millennials, social constructionism, membership-categorization devices, categorybound activities, communication rules, organizational assimilation

Mémoire numérisé par la Direction des bibliothèques de l'Université de Montréal.

Generalization error describes how well a supervised machine learning algorithm predicts the labels of input data that it has not been trained with. This project aims to explore two different methods for bounding generalization error, f-CMI and ISMI, which explicitly use mutual information. Our experiments are based on the experiments in the papers in which the methods were proposed. The experiments implement and validate the accuracy of the mathematically derived bounds. Each methodology also has a different method for calculating mutual information. The ISMI bound experiment used a multivariate normal distribution dataset, whereas a dataset consisting of cats and dogs was used for the experiment using f-CMI. Our results show that both methods are capable of bounding the generalization error of a binary classification algorithm and provide bounds that closely follow the true generalization error. The results of the experiments agree with the original experiments, indicating that the proposed methods also work for similar applications with different datasets. / Generaliseringsfel beskriver hur väl en övervakad maskininlärnings algoritm förutspår etiketter av indata som den inte har blivit tränad med. Syftet med projektet är att utforska två olika metoder för att begränsa generaliseringsfelet, f-CMI och ISMI som explicit använder ömsesidig information. Vårt experiment är baserat på experimenten i artiklarna som tog fram metoderna. Experimenten implementerade och validerade noggrannheten av de matematiskt härleda gränserna. Varje metod har olika sätt att beräkna den ömsesidiga informationen. ISMI gräns experimentet använde en flerdimensionell normalfördelning som data set, medan en datauppsättning med katter och hundar användes för f-CMI gränsen. Våra resultat visar att båda metoder kan begränsa generaliseringsfelet av en binär klassificerings algoritm och förse gränser som nära följer det sanna generaliseringsfelet. Resultatet av experimenten instämmer med de ursprungliga författarnas experiment vilket indikerar att de föreslagna metoderna också fungerar for liknande tillämpningar med andra data set. / Kandidatexjobb i elektroteknik 2022, KTH, Stockholm

Nonlinear explicit finite element (FE) simulations are used to study the axial collapse behavior of multi-corner. single- and multi-cell crush tubes under quasi-static and dynamic loading conditions. It is shown that the higher hardening modulus and yield stress increases the crush force and its resulting energy absorption. Moreover, the multi-cell tubes are found to have complicated collapse modes because of the geometrical complexity of the corner region unlike single-cell tubes. it was also shown that the stress wave propagation has a significant effect on the formation of crush modes in the tubes without imperfections whereas this effect can be ignored in tubes with imperfection or trigger mechanism. An analytical formula for the prediction of mean crush force of multi-corner multi-cell tubes is derived based on the super folding element theory. The analytical predictions for the mean crush force are found to be in good agreement with the FE solutions. Results also show a strong correlation between the cross-sectional geometry and the crash behavior with the method of connecting the inner to the outer walls having large influence on the energy absorption.

No description available.

No description available.

No description available.

No description available.

No description available.

People unconsciously learn spatial information about places they encounter frequently, leading them to search through familiar scenes faster than for unfamiliar scenes. We explored this phenomenon—the contextual cueing effect—in scenes containing images of different human faces. Participants searched through a series of scenes for a target among distractors, characterized as a letter T among letter L’s with each letter positioned on top of a face image (Experiment 1) or as a female face among male faces (Experiment 2). Experiment 1 showed that when the binding of identity and location was manipulated during learning, slightly greater (but not statistically significant) contextual cueing effects were found for repeated scenes with constant identity-to-location binding than those repeated scenes with constant spatial configurations but shuffled identity-to-location binding. Experiment 2 showed that if the binding of identity-to-location changed after the learning of a set of identity-to-location binding, small (but not statistically significant) costs of contextual cueing were found. The results suggest that in the contextual cueing paradigm, repeated identity-to-location binding might be learned but the learning of repeated spatial configurations alone account for a major portion of the learning. / Thesis / Master of Science (MSc)