13594 research outputs found
Sort by
Choosing to Work: The Impact of Gender Role Attitudes on Female Labor Force Participation in South Korea
No full textWhile female labor force participation in South Korea has increased with industrialization, it has not kept pace with the rates of other OECD countries. A potential factor that may explain the labor force participation decision amongst females in South Korea is a female’s attitudes towards gender roles. Using 2012 data from the Korean General Social Survey (KGSS), I find that more modern attitudes increase the likelihood of participating in the labor force, particularly for married women. Over the full sample, a more modern attitude increases the likelihood of participating in the labor force more than pursuing an undergraduate college education beyond high school, but these impacts differ for married and unmarried women when they are considered separately. The results suggest that policies to change public attitudes surrounding women and work, implemented alongside preexisting policies such as childcare support, could further increase female labor force participation. Raising the female labor force participation rate (FLPR) could help mitigate the consequences of the demographic crisis that South Korea is currently facing and set it on a path towards more sustainable economic growth
The Effects of Prometheus Bound on China During the Early Twentieth Century and Rebel Plays in South Africa
No full textEarly Modern Continental Philosophy
No full textA close analytical reading of selected texts from the classical Continental rationalists,with a focus on epistemology and metaphysics. Topics include: ideas, skepticism, belief, knowledge, science, bodies, minds, God, causation, natural laws, afterlife, personal identity, and free will. \ud
Our main figures include Descartes, Malebranche, Spinoza, and Leibniz
Abnormal Psychology
No full textThis course surveys the major forms of emotional and behavioral disorders including their definition, etiology, and treatment. We will use an integrative approach, drawing from biological, psychological, and cultural models of psychopathology. Special attention will be paid to research methods used to develop models of psychopathology, their accurate interpretation, and the use of empirically supported treatments to address emotional distress
Auditing Deep Neural Networks and Other Black-box Models
No full textIn this era of self-driving cars, smart watches, and voice-commanded speakers, machine learning is ubiquitous. Recently, deep learning has shown impressive success in solving many machine learning problems related to image data and sequential data - with the result that people are frequently impacted by deep learning models on a daily basis. However, how do we judge if these models are fair, and how do we discover what information is important when making a decision? And as APIs become ever-more common, how do we determine this information if we do not have access to the model itself? We developed a novel technique called Gradient Feature Auditing which gradually obscures information from a data-set and evaluates how a model's predictions change as yet more of that information is obscured. This allows a deeper investigation of what information and features are actually used by machine learning models when making predictions. Throughout our experiments, we apply Gradient Feature Auditing on multiple data-sets using several popular modeling techniques (linear SVMs, C4.5 decision trees, and shallow feed-forward neural networks) to provide evidence that Gradient Feature Auditing indeed affords deeper insight into what information a model is using
Using Phrase-Structure Rules in Extractive Text Summarization
No full textDocument summarization has become increasingly important since the explosion of data creation that began in the mid-20th century. Improving automated text summarization is essential in order to provide methods of generating summaries either for the purpose of replacing a larger document or for indexing. The most common form of automatic text summarization utilizes extractive techniques, which nearly always rely on orthographic sentences as the main textual unit. By using phrase-structure sentences instead, we are able to isolate meaning and importance at the sub-sentence, but super-word, level. In changing the primary textual unit from orthographic sentence to phrase-structure sentence, we hope to see a marked improvement in the quality of generated abstracts, as assessed by the ROUGE evaluation system
Optimizing a Machine Learning System for Materials Discovery
No full textAdvanced functional materials are crucial for addressing numerous challenges in medicine, communications, and energy. As highlighted by the White House Materials Genome Initiative, computational tools are critical for improving the materials discovery process. Many of the most promising materials are inorganicorganic hybrid materials. This broad class of compounds exhibits extraordinary structural diversity that has made them a topic of interest as materials for applications including energy storage, catalysis, photovoltaics, optical engineering, and gas sorption. The Norquist lab focuses on the exploratory synthesis of organically-templated metal oxides. While computational techniques have been applied extensively to predicting material properties, the Dark Reactions Project (DRP) takes a different, underexplored, approach using machine learning to improve the synthesis process itself. This system is already in use in the Norquist lab, increasing reaction success rates, but further improvement could make it a more reliable source of new reaction suggestions in unexplored chemical space. This thesis enhances the DRP by constructing machine learning models of several types and using them to investigate different ways of describing a chemical reaction. It investigates both automated feature selection and varied descriptors based on reasoned changes and variations of model cost functions to make them more reflective of the actual use case. Many models in this work improve on the previous models, and the best achieves an average accuracy of 80% predicting unseen reactions. The Matthews coeffi cient, a more robust measure of performance which indicates the correlation between predicted and actual outcomes, increases from 0.25 to 0.43. When incorporated into the recommendation pipeline, these models should result in improved reaction recommendations from the DRP and chemical hypotheses
Auditing Deep Neural Networks to Understand Recidivism Predictions
No full textIn recent years, deep neural network models have proven to be incredibly accurate on many classification benchmarks. Due to this high accuracy, many non-technical fields are interested in using these models to assist in decision making processes. However, this curiosity is generally tempered by the realization that it is di fficult to understand what features of the data contribute to the prediction. We present a method to evaluate the effect of each feature in a data set on the predictions of a model, which we refer to as gradient feature auditing (GFA). To test this method, we trained four models (a deep neural network, SVM, SLIM, and decision tree) on recidivism data and then applied GFA to each model. The experimental portion verified the ability of GFA to obtain a ranked ordering of features. Next, we attempted to use methods from interpretable learning to validate our procedure. Overall, GFA allows domain experts to use the most effective model of their data in the decision making process, while also retaining the ability to explain how those decisions are being made
Automatic Emotion Detection Technology for Autism Therapy
No full textEmotion recognition technology is being used to provide feedback to advertisers, sharing emotions through social media, and providing assistance to those who struggle to detect emotions. Affectiva is one of the companies whose emotion detection software is being used for these purposes. People with autism have trouble detecting emotion in animated faces and humans. The Affdex SDK provided by Affectiva gives developers a chance to use Affectiva’s technology for their own ideas and purposes. Affdex provides a fast and accurate method for facial feature tracking and emotion output. The Affdex SDK can be used to create a game for autism therapy
Of Monsters and Marvels
No full textFrom acts of devotion to encounters with the strange and the monstrous, this course examines the place of wonder in Islamic traditions through readings from the Qur’an, exegesis, and prophetic traditions;literary encounters with the Arabian Nights and the Shah-nameh; travel narratives, descriptive geography, and cosmography; philosophy and theology. In addition to literary sources, we will draw\ud
on visual media through examples from the use of calligraphy, illuminated manuscripts, cartographical projections, and architectural monuments. Topics include the role of the sublime in wonder traditions (ajā’ib); conceptions of natural/unnatural orders; the discourses of alterity; the theodicy of divine design; projections of space in realms of the sacred and the profane