Tesi sul tema "Bias"

This study examined a feature of organizations that, prior to this work had not been identified or defined: Organizational Bias. If an organization can learn, have an identity and memory, then an organization can hold bias. Organizational bias occurs when an organization’s culture, as defined and reproduced within its interactions with agents and actors, prevents actors from becoming agents by denying them the power required to change the organizational structure. This exclusionary aspect of bias creates a significant obstacle for educational institutions, many of whom define their missions as serving all students and providing a place that builds, supports, and serves community. Organizational ethnography was used as the methodology to examine organizational bias. Data were collected in a committee space designed to be inclusive of community members and other actor voices as they evaluated new high school course proposals. Data were comprised of the following: observations of all meetings, documents, such as correspondence between members and documents produced as a result of or that informed the committee’s work, and interviews with members. The data were analyzed using a combination of interactional analysis and axial coding. Findings illustrated that patterns of exclusion initiated through the institutional structure of talk constituted and reconstituted organizational bias. This structure created relevant opportunities for resource use (e.g., credit information), which agents were able to perform, but from which actors were excluded, that afforded them the agency to reconstitute the practical and tacit knowing of the organization, which then reproduced the initial structure. Implications for this work include a clearer understanding of how educational organizations hold bias, what patterns of interaction to examine, and how to interrupt the reconstitution of those practices to be more inclusive of actors in an effort to work more closely toward the defined mission.

Sensor fusion has long been recognized as a mean to improve target tracking. Sensor fusion deals with the merging of several signals into one to get a better and more reliable result. To get an improved and more reliable result you have to trust the incoming data to be correct and not contain unknown systematic errors. This thesis tries to find and estimate the size of the systematic errors that appear when we have a multi platform environment and data is shared among the units. To be more precise, the error estimated within the scope of this thesis appears when platforms cannot determine their positions correctly and share target tracking data with their own corrupted position as a basis for determining the target's position. The algorithms developed in this thesis use the Kalman filter theory, including the extended Kalman filter and the information filter, to estimate the platform location bias error. Three algorithms are developed with satisfying result. Depending on time constraints and computational demands either one of the algorithms could be preferred.

Public reason theorists argue that it is permissible for the state to enforce political norms, such as laws or constitutional principles, when those norms are acceptable to ?reasonable people.? Reasonable people are neither actual people, with all their flaws, nor are they perfect people; they are rather a partially idealized group ? realistic in some ways and idealized in others. Each of the major public reason theorists ? John Rawls, Gerald Gaus, Jonathan Quong, Joshua Cohen ? idealizes reasonable people to a different degree, but they all share two claims: 1) Reasonable people hold diverse views of the good life. Nevertheless, 2) Reasonable people can all accept basic liberal political norms grounded in freedom and equality. My dissertation begins by arguing that theorists are not free to choose any level of idealization, but are constrained in this choice by the justifications of their theories. In particular, idealization is constrained by one essential part of public reason?s justification, which I call the ?diversity argument.? The diversity argument explains the first element of reasonable people: why do they disagree about the good? The answers, I argue, attributes certain realistic qualities and tendencies to reasonable people, which therefore constrains how much we can idealize them. In chapters on the major public reason theorists, I argue that they all offer a diversity argument that does not match the level of idealization that they employ. As a result, they are unable to show that liberal norms are acceptable to reasonable people, appropriately idealized. In the final chapter, I argue that the mismatch in these theories goes even deeper, which we can see when we ask why we must accommodate disagreement at all. The answers that philosophers have given us ? reasonable disagreement is the inevitable result of human reasoning, human psychology, or free conditions ? also apply to irrational disagreement. Irrational influences such as implicit bias and motivated reasoning are also inevitable results of who we are and how we live, which means we must accommodate these realistic tendencies in political justification. So, if public reason theories must now accommodate disagreement among reasonable-but-sometimes-irrational people, what could be acceptable to all such people? I conclude by suggesting a new direction for public reason theories. People who disagree about the good life, but recognize their common biases, can still justify their views to each other by supporting institutions that mitigate those biases, such as non-discrimination laws and deliberative institutions. This requires a new kind of social contract theory ? one that is grounded in the shared recognition of our limitations, rather than our shared reasons.

This dissertation explores the interaction of statistical learning and bias. While novel category learning is known to involve statistical learning mechanisms, there is clear evidence that learners bring bias of various forms into the process, resulting in learning patterns that differ from the expected outcome of purely distributional learning. Here, I explore the way that the learner's experienced-based expectations about the informativeness of acoustic cues interacts with the informativeness of those cues during novel category learning. This question is addressed using a combination of experimental and computational methods. These two sources provide complementary evidence about the role of learning bias in statistical category learning, examining the effects of bias under certain learning conditions and testing one way in which such bias might be computationally implemented in the learning process.

Chapter 3 presents experimental evidence that learners are sensitive to gradient differences in cue reliability and demonstrate evidence of using multiple cues in distributional learning, contrasting with previous results. Three groups of native English speaker participants were trained to distinguish a pair of novel vowel categories with cues which are, for them, either uninformative at the segmental level (F0) or highly informative for native vowel distinctions (F2), where the relative informativeness of the cues was varied between groups. Participants in groups where one of the cues was more informative showed cue weightings which were influenced by the weighting present in the stimuli, demonstrating that learners' cue weighting is modulated by the properties of the input. The balanced-cues group, more consistent with previous studies, showed less reliable learning. All groups showed a cue weighting pattern which is not directly attributable to L1 experience, preferring F0-based categorization boundaries over F2-based boundaries.

Chapter 4 presents a novel model of category learning which examines the interaction of bias and statistical learning from a computational perspective. This model implements category learning as a form of hierarchical inference where the learner makes inferences at the local level, learning individual categories, but also at a global level, making inferences about the expected properties of unobserved categories. The learner generalizes about cue informativeness in their L1. This influences what the learner expects cue weightings in novel vowel categories to look like. In a series of simulations, the model's ability to learn L1 categories and, subsequently, novel categories after L1 training, is tested and compared to a model which does not implement learning bias. These simulations test whether the presence of bias is beneficial in L1 learning, and how the bias developed over the L1 learning process influences later learning of novel categories.

The results of the simulations offer a potential explanation of the bias observed in the experimental results presented here, and suggests a number of avenues for future research. Notably, a number of other distributional learning results in previous research might also be amenable to the same explanation proposed here. This suggests that difficulties in multidimensional distributional learning for adults might arise more from issues of experimental and stimulus design than inherent difficulty of the task.