This master thesis project researches the learning preferences of external clients of Ericsson Learning Services Department, focusing on adult corporate training. The purpose is to collect insights and opinions of clients and use their answers as a basis of improvement for updating and developing courses in the future. The project employed a mix method approach to answer the research question, which combined qualitative data collection through interviews and surveys for a thematic analysis and quantitative data collection from former survey results for statistical analyses. Furthermore, a literary study was conducted to support the research process.  The main results from the thematic analysis proved that most external clients preferred an online learning setting or face to face training. Online courses were mostly preferred due to convenience, even if most would agree that a classroom setting is the best setting for optimal learning. Results from the statistical analysis gave few valuable insights. This may depend on how the questions in the survey were asked and therefore the main result of the statistical analysis is how to improve the surveys using the Learning Experience Questionnaire (LEQ).  This project opens possibilities for applications and future research in this area. One suggestion is to conduct a follow-up study where the surveys are updated and tailored to the needs and wants of the external clients which enables a more comprehensive analysis. Furthermore, researching the specific elements of classroom-based learning that external clients find valuable could provide more understanding to further better the online learning experience. These paths of further research can contribute to the ongoing improvement of adult corporate learning. / Detta examensarbete undersöker lärandepreferenser hos externa kunder till Ericssons Learning Services Department och fokuserar på vuxenutbildning på företag. Målet är att samla insikter och åsikter från externa kunder och använda deras svar för ett förbättringsunderlag för uppdatering och utveckling av kurser. Examensarbetet använde en blandning av metoder för att besvara frågeställningarna, vilket kombinerade kvalitativ datainsamling via intervjuer och frågeformulär till en tematisk analys med kvantitativ datainsamling från äldre undersökningsresultat till statistiska analyser. Vidare genomfördes en litteraturundersökning som stöd i undersökningsprocessen.  De huvudsakliga resultaten från den tematiska analysen visade att de flesta externa kunder föredrog en lärandemiljö online eller traditionell klassrumsundervisning med den gemensamma nämnaren att kursen är lärarledd. Onlinekurser föredrogs mest på grund av bekvämligheten, även om de flesta skulle hålla med om att en klassrumsmiljö är den bästa miljön för optimalt lärande. Resultaten från den statistiska analysen gav få värdefulla insikter. Detta kan bero på hur frågorna i undersökningen ställs och därför är huvudresultatet av den statistiska analysen hur man kan förbättra undersökningarna med hjälp av Learning Experience Questionnaire (LEQ). Detta undersökningsresultat öppnar upp möjligheter till vidare forskning på området. Ett förslag är att genomföra en uppföljande studie då frågeformulären är uppdaterade och skräddarsydda till externa kunders behov och önskemål som möjliggör en mer omfattande analys. Vidare kan undersökning av de specifika element som kunder uppskattar med klassrumsbaserat lärande för att ge insikter om hur man kan optimera upplevelser av lärande online. Dessa vägar av vidare forskning kan bidra till den pågående förbättringen av vuxenutbildning på företag.

This dissertation used three independent studies to evaluate learning adaptations, techniques, and tools in response to COVID-19. Study 1 provides a personal narrative about COVID-10 and school closures. The first-hand reflection explores technologies used for remote education and how the technologies can improve learning experiences. Study 2 is a literature review identifying the challenges faced in providing engaging lessons and overcoming technology fatigue. Study 3 is an empirical study providing insight into the continued use of video instruction and virtual simulations. Collectively, the manuscripts in this dissertation provide the factors influencing the normalization of digital tools used during remote learning. The technology adoption cycle and the normalization process theory are used to discuss factors that influence the normalization of tools and techniques post-pandemic. The findings of this dissertation provide insight into future uses of digital tools and techniques, given the lessons learned from emergency remote learning due to the COVID-19 pandemic.

Using a modified still-face procedure the present study explores 3-6-month-old infants' behavioral and physiological responses to mothers' screen distractions during mother-infant interactions. In the modified phone still-face procedure the neutral face of the traditional still face procedure was replaced with mothers' texting on their mobile phones. Infants' cortisol stress responses to mothers' device use were assessed through the collection of 3 infant saliva samples. Infants' behavioral responses including facial expressions, vocalizations, gaze and self-comforting behaviors were also explored. All mother-infant interactions were videoed recorded and coded for analysis. Thirty-four mother-infant dyads participated, average ages for mothers was 29 years and 4.4 months for infants. As predicted, infants demonstrated the changes in affect associated with the still-face effect, with significant differences in positive and negative affect during the play phases and the phone still face phase. As a whole, infants did not respond with increased cortisol responses, however, when individual differences were explored 47% responded with increased stress during mothers' phone distractions. Mother's frequency and attitudes towards device use were also assessed but were unrelated to infant responses. Implications and directions for future research are discussed.

Call number: LD2668 .T4 1985 S53 / Master of Science-

This research addresses the novel application of design of experiment, artificial neural net-works and logistic regression to study the effect of demographic characteristics on the risk of acquiring HIV infection among the antenatal clinic attendees in South Africa. The annual antenatal HIV survey is the only major national indicator for HIV prevalence in South Africa. This is a vital technique to understand the changes in the HIV epidemic over time. The annual antenatal clinic data contains the following demographic characteristics for each pregnant woman; age (herein called mother's age), partner's age (herein father's age), population group (race), level of education, gravidity (number of pregnancies), parity (number of children born), HIV and syphilis status. This project applied a screening design of experiment technique to rank the effects of individual demographic characteristics on the risk of acquiring an HIV infection. There are a various screening design techniques such as fractional or full factorial and Plackett-Burman designs. In this work, a two-level fractional factorial design was selected for the purposes of screening. In addition to screening designs, this project employed response surface methodologies (RSM) to estimate interaction and quadratic effects of demographic characteristics using a central composite face-centered and a Box-Behnken design. Furthermore, this research presents the novel application of multi-layer perceptron’s (MLP) neural networks to model the demographic characteristics of antenatal clinic attendees. A review report was produced to study the application of neural networks to modelling HIV/AIDS around the world. The latter report is important to enhance our understanding of the extent to which neural networks have been applied to study the HIV/AIDS pandemic. Finally, a binary logistic regression technique was employed to benchmark the results obtained by the design of experiments and neural networks methodologies. The two-level fractional factorial design demonstrated that HIV prevalence was highly sensitive to changes in the mother's age (15-55 years) and level of her education (Grades 0-13). The central composite face centered and Box-Behnken designs employed to study the individual and interaction effects of demographic characteristics on the spread of HIV in South Africa, demonstrated that HIV status of an antenatal clinic attendee was highly sensitive to changes in pregnant mother's age and her educational level. In addition, the interaction of the mother's age with other demographic characteristics was also found to be an important determinant of the risk of acquiring an HIV infection. Furthermore, the central composite face centered and Box-Behnken designs illustrated that, individual-ally the pregnant mother's parity and her partner's age had no marked effect on her HIV status. However, the pregnant woman’s parity and her male partner’s age did show marked effects on her HIV status in “two way interactions with other demographic characteristics”. The multilayer perceptron (MLP) sensitivity test also showed that the age of the pregnant woman had the greatest effect on the risk of acquiring an HIV infection, while her gravidity and syphilis status had the lowest effects. The outcome of the MLP modelling produced the same results obtained by the screening and response surface methodologies. The binary logistic regression technique was compared with a Box-Behnken design to further elucidate the differential effects of demographic characteristics on the risk of acquiring HIV amongst pregnant women. The two methodologies indicated that the age of the pregnant woman and her level of education had the most profound effects on her risk of acquiring an HIV infection. To facilitate the comparison of the performance of the classifiers used in this study, a receiver operating characteristics (ROC) curve was applied. Theoretically, an ROC analysis provides tools to select optimal models and to discard suboptimal ones independent from the cost context or the classification distribution. SAS Enterprise MinerTM was employed to develop the required receiver-of-characteristics (ROC) curves. To validate the results obtained by the above classification methodologies, a credit scoring add-on in SAS Enterprise MinerTM was used to build binary target scorecards comprised of HIV positive and negative datasets for probability determination. The process involved grouping variables using weights-of-evidence (WOE), prior to performing a logistic regression to produce predicted probabilities. The process of creating bins for the scorecard enables the study of the inherent relationship between demographic characteristics and an in-dividual’s HIV status. This technique increases the understanding of the risk ranking ability of the scorecard method, while offering an added advantage of being predictive.

Master of Science / Department of Psychology / Lester C. Loschky / To investigate the role of various bands of spatial frequencies for drawing, untrained artists drew four portraits from four different bands of spatial frequencies (e.g. unfiltered, 4-8, 8-16, & 16-32 cycles per face width (c/fw)). Raters then judged the accuracy of the drawings in comparison to both the source image from which the drawings were produced and an unfiltered version of the same face. The results show that low spatial frequencies (LSFs) and high spatial frequencies (HSFs) were useful for drawing, relative to middle spatial frequencies (MSFs). Additionally, the unfiltered condition that contained all spatial frequencies produced the most accurate drawings. This suggests that when artists are allowed access to both LSFs and HSFs they are able to utilize the global structure information carried in LSFs as well as the edge and detail information carried in HSFs to create more accurate drawings. The author posits that the MSFs that are useful for face recognition become redundant for drawing and that novice artists discount these MSFs in the control condition in order to increase the saliency and usefulness of the LSFs and HSFs. The results have implications for art education, drawing technology and the development of low-level drawing theory.

This research addresses the novel application of design of experiment, artificial neural net-works and logistic regression to study the effect of demographic characteristics on the risk of acquiring HIV infection among the antenatal clinic attendees in South Africa. The annual antenatal HIV survey is the only major national indicator for HIV prevalence in South Africa. This is a vital technique to understand the changes in the HIV epidemic over time. The annual antenatal clinic data contains the following demographic characteristics for each pregnant woman; age (herein called mother's age), partner's age (herein father's age), population group (race), level of education, gravidity (number of pregnancies), parity (number of children born), HIV and syphilis status. This project applied a screening design of experiment technique to rank the effects of individual demographic characteristics on the risk of acquiring an HIV infection. There are a various screening design techniques such as fractional or full factorial and Plackett-Burman designs. In this work, a two-level fractional factorial design was selected for the purposes of screening. In addition to screening designs, this project employed response surface methodologies (RSM) to estimate interaction and quadratic effects of demographic characteristics using a central composite face-centered and a Box-Behnken design. Furthermore, this research presents the novel application of multi-layer perceptron’s (MLP) neural networks to model the demographic characteristics of antenatal clinic attendees. A review report was produced to study the application of neural networks to modelling HIV/AIDS around the world. The latter report is important to enhance our understanding of the extent to which neural networks have been applied to study the HIV/AIDS pandemic. Finally, a binary logistic regression technique was employed to benchmark the results obtained by the design of experiments and neural networks methodologies. The two-level fractional factorial design demonstrated that HIV prevalence was highly sensitive to changes in the mother's age (15-55 years) and level of her education (Grades 0-13). The central composite face centered and Box-Behnken designs employed to study the individual and interaction effects of demographic characteristics on the spread of HIV in South Africa, demonstrated that HIV status of an antenatal clinic attendee was highly sensitive to changes in pregnant mother's age and her educational level. In addition, the interaction of the mother's age with other demographic characteristics was also found to be an important determinant of the risk of acquiring an HIV infection. Furthermore, the central composite face centered and Box-Behnken designs illustrated that, individual-ally the pregnant mother's parity and her partner's age had no marked effect on her HIV status. However, the pregnant woman’s parity and her male partner’s age did show marked effects on her HIV status in “two way interactions with other demographic characteristics”. The multilayer perceptron (MLP) sensitivity test also showed that the age of the pregnant woman had the greatest effect on the risk of acquiring an HIV infection, while her gravidity and syphilis status had the lowest effects. The outcome of the MLP modelling produced the same results obtained by the screening and response surface methodologies. The binary logistic regression technique was compared with a Box-Behnken design to further elucidate the differential effects of demographic characteristics on the risk of acquiring HIV amongst pregnant women. The two methodologies indicated that the age of the pregnant woman and her level of education had the most profound effects on her risk of acquiring an HIV infection. To facilitate the comparison of the performance of the classifiers used in this study, a receiver operating characteristics (ROC) curve was applied. Theoretically, an ROC analysis provides tools to select optimal models and to discard suboptimal ones independent from the cost context or the classification distribution. SAS Enterprise MinerTM was employed to develop the required receiver-of-characteristics (ROC) curves. To validate the results obtained by the above classification methodologies, a credit scoring add-on in SAS Enterprise MinerTM was used to build binary target scorecards comprised of HIV positive and negative datasets for probability determination. The process involved grouping variables using weights-of-evidence (WOE), prior to performing a logistic regression to produce predicted probabilities. The process of creating bins for the scorecard enables the study of the inherent relationship between demographic characteristics and an in-dividual’s HIV status. This technique increases the understanding of the risk ranking ability of the scorecard method, while offering an added advantage of being predictive.

La téléphonie mobile se démocratise et de nouveaux types de réseaux voient le jour, notamment les réseaux ad hoc. Sans focaliser exclusivement sur ces réseaux particuliers, le nombre de communications vocales effectuées chaque minute est en constante augmentation mais les réseaux sont encore souvent victimes d'erreurs de transmission. L'objectif de cette thèse porte sur l'utilisation de méthodes de codage en vue d'une transmission de la voix robuste face aux pertes de paquets, sur un réseau mobile et sans fil perturbé permettant le multichemin. La méthode envisagée prévoit l'utilisation d'un codage en descriptions multiples (MDC) appliqué à un flux de données issu d'un codec de parole bas débit, plus particulièrement l'AMR-WB (Adaptive Multi Rate - Wide Band). Parmi les paramètres encodés par l'AMR-WB, les coefficients de la prédiction linéaire sont calculés une fois par trame, contrairement aux autres paramètres qui sont calculés quatre fois. La problématique majeure réside dans la création adéquate de descriptions pour les paramètres de prédiction linéaire. La méthode retenue applique une quantification vectorielle conjuguée à quatre descriptions. Pour diminuer la complexité durant la recherche, le processus est épaulé d'un préclassificateur qui effectue une recherche localisée dans le dictionnaire complet selon la position d'un vecteur d'entrée. L'application du modèle de MDC à des signaux de parole montre que l'utilisation de quatre descriptions permet de meilleurs résultats lorsque le réseau est sujet à des pertes de paquets. Une optimisation de la communication entre le routage et le processus de création de descriptions mène à l'utilisation d'une méthode adaptative du codage en descriptions. Les travaux de cette thèse visaient la retranscription d'un signal de parole de qualité, avec une optimisation adéquate des ressources de stockage, de la complexité et des calculs. La méthode adaptative de MDC rencontre ces attentes et s'avère très robuste dans un contexte de perte de paquets.

L'intolérance à l'incertitude (II) et l'attitude négative face aux problèmes (AN) sont des vulnérabilités cognitives fortement liées à la tendance excessive à s'inquiéter et à d'autres symptômes du trouble d'anxiété généralisée (TAG; Dugas, Gagnon, Ladouceur, & freston, 1998). Tel que proposé par Dugas, Buhr et Ladouceur (2004), certains contextes familiaux pourraient être liés au développement de ces vulnérabilités cognitives. Des études ont démontré un lien entre des comportements parentaux d'anxiété et de surprotection et des symptômes d'anxiété chez l'enfant (Bôgels & Brechman-Toussaint, 2006) et l'adolescent (van Brakel, Muris, Bôgels, & Thomassen, 2006). Une méta-analyse a conclu que les adultes anxieux auraient perçu un patron de comportements parentaux surprotecteurs et peu chaleureux durant l'enfance (Gerlsma, Emmelkamp, & Arrindell, 1990). Malgré l'abondance d'écrits portant sur l'anxiété et les pratiques parentales, peu d'études ont clarifié les comportements parentaux liés spécifiquement au diagnostic de TAG et de ses caractéristiques. La présente étude propose d'examiner les liens entre les comportements parentaux, les symptômes du TAG, l'II et l'AN, en comparant deux groupes de participants : un groupe présentant les critères diagnostiques du TAG et un groupe de participants appariés en bonne santé psychologique (groupe contrôle). L'hypothèse 1 proposait que les adultes souffrant d'un TAG rapporteraient avoir perçu davantage de comportements parentaux d'anxiété et de surprotection que leurs homologues contrôles. L'hypothèse 2 stipulait que I'll et l'AN seraient chacune liée à ces comportements parentaux. Advenant la présence de liens significatifs entre ces variables, la thèse visait à vérifier le rôle médiateur de l'II et de l'AN dans les liens unissant les comportements parentaux et les symptômes du TAG. Les résultats révèlent que les individus ayant un TAG ont perçu davantage de comportements anxieux et surprotecteurs chez leurs deux parents ainsi que moins de chaleur maternelle que leurs homologues non anxieux. Le rôle médiateur de l'II et de l'AN a été confirmé entre, d'une part, la perception de comportements anxieux et surprotecteurs du père, et d'autre part, les symptômes du TAG. L'II apparaît aussi être une variable médiatrice dans le lien unissant la perception de comportements de chaleur paternelle et les symptômes du TAG. L'II et de l'AN constituent des variables médiatrices entre la perception de comportements de chaleur maternelle et les symptômes du TAG, alors que l'AN exerce un rôle médiateur entre la surprotection maternelle et les symptômes du TAG. Ces résultats appuient ceux de Zlomke et Young (2009), qui ont récemment suggéré le rôle médiateur de t'II dans la relation entre les comportements anxieux des parents et l'inquiétude auprès d'étudiants universitaires. La présente étude est la première à illustrer le rôle médiateur de l'll et de l'AN dans la relation entre les comportements parentaux perçus durant l'enfance et les symptômes du TAG et ce, auprès d'un échantillon clinique adulte. Les résultats informent quant aux mécanismes liés au TAG et suggèrent la pertinence d'interventions précoces ciblant ces vulnérabilités cognitives auprès des enfants anxieux et de leur famille.

In this dissertation, we focus on solving Nonlinear Least Squares problems using a supervised approach. In particular, we developed a Supervised Descent Method (SDM), performed thorough theoretical analysis, and demonstrated its effectiveness on optimizing analytic functions, and four other real-world applications: Inverse Kinematics, Rigid Tracking, Face Alignment (frontal and multi-view), and 3D Object Pose Estimation. In Rigid Tracking, SDM was able to take advantage of more robust features, such as, HoG and SIFT. Those non-differentiable image features were out of consideration of previous work because they relied on gradient-based methods for optimization. In Inverse Kinematics where we minimize a non-convex function, SDM achieved significantly better convergence than gradient-based approaches. In Face Alignment, SDM achieved state-of-the-arts results. Moreover, it was extremely computationally efficient, which makes it applicable for many mobile applications. In addition, we provided a unified view of several popular methods including SDM on sequential prediction, and reformulated them as a sequence of function compositions. Finally, we suggested some future research directions on SDM and sequential prediction.