18736 research outputs found
Sort by
A Symbolic Approach to Nonlinear Time Series Analysis
Get PDFCurrent nonlinear time series methods such as neural networks forecast well. However, they act as a black box and are difficult to interpret, leaving the researchers and the audience with little insight into why the forecasts are the way they are. There is a need for a method that forecasts accurately while also being easy to interpret. This paper aims to develop a method to build an interpretable model for univariate and multivariate nonlinear time series data using wavelets and symbolic regression. The final method relies on multilayer perceptron (MLP) neural networks as a form of dimensionality reduction and the PySR algorithm to determine the symbolic relationships. It also explores use cases for using the discrete wavelet transformation to extract information from the dataset
Game UI/UX Design with Original Character Concept Design and Dynamic Effects
Get PDFFor my thesis project, I selected three masteries: Character Concept Design, Dynamic Effects, and UI/UX Design. For my UI/UX design, I incorporated diverse character designs for player selection, leading to character concept design as a secondary focus. Additionally, integrating dynamic effects aimed to enhance the visual appeal of the project and further highlight the character\u27s personality
A 1.1GS/s 12-bit single-channel pipelined-SAR ADC with an improved CDAC implementation and adaptive interstage redundancy
No full textThis dissertation presents a 12-bit 1.1 GS/s single-channel pipelined-SAR ADC implemented in a 28nm CMOS technology. A new technique that provides adaptive inter-stage redundancy is proposed to mitigate the speed overhead of the conventional inter-stage redundancy bit. In addition, the first-stage CDAC is implemented with a large-DAC and a small-DAC to improve the settling speed during the bit conversions, and a high-speed detect-and-skip decoder with minimum power overhead is incorporated to reduce the switching power without affecting the high-speed operation. The adaptive inter-stage redundancy improves the speed of the third stage by 10%. The CDAC1 improves the speed of the first stage by 11% and reduces the switching power on large-DAC by 25%. Overall, the single-channel ADC achieves an SNDR of 60.1 dB and an SFDR of 75.3 dB at Nyquist input operating at 1.1 GS/s. With 8.5mW power consumption at a 0.9 V power supply, it achieves a Walden FOM of 9.3 and a Schreier FOM of 168.2 dB
Parsing the Paramour: How Players’ Sexuality, Motivation, and UDO Influence Their Romantic Subplot Decisions in Western RPGs
Get PDFThis article is the culmination of a study conducted by the author on players of five Western Role-playing Games (WRPGs) who participated in the romantic subplots of those games (n=1001). Influenced by Sherry Turkle’s The Second Self (2005) and publications from Adams (2015), Dym (2019), and McDonald (2015), this study investigated the relationship between participants’ sexualities, player motivation types, and universal diversity orientation (UDO) scores. This study found data to support significant relationships between these variables as well as (a) the frequency with which player sexuality influences their in-game romantic decisions, (b) the perception of how different sexualities are portrayed in WRPGs, and (c) the perception of how diverse sexualities fit along the games’ critical narrative paths. These results contribute to the body of work that advises the industry to spend the energy and resources on more inclusive narratives and better portrayals of the sexuality spectrum
Corporate Human Trafficking
Get PDFThe utilization of the internet for human trafficking and sexual exploitation is not an issue that can be tackled one corporation, one country, or one market sector at a time. It is an international problem that requires broader solutions that can protect and provide remedy to victims without chilling the freedom of speech and freedom of contract of consensual parties engaged in sex work. Recent changes to laws related to human trafficking have strengthened the power of litigation, authorizing civil lawsuits against perpetrators of human trafficking that may include third-parties who knowingly benefit from trafficking conduct—such as internet providers, business partners and even banks and credit card companies. These laws have enabled the victims of Jeffrey Epstein to successfully pursue Deutsche Bank and JP Morgan Chase, receiving multimillion dollar settlements. Pressure from credit card companies who were named in lawsuits combined with other litigation efforts changed the practices of Pornhub and its parent company MindGeek, resulting in the eventual acquisition of MindGeek by Ethical Capital Partners (ECP), a private equity firm intent on giving the company an Environmental, Social, Governance (ESG) makeover. While there is a next chapter for the parent corporation, many of the independent sex workers who depend on platforms for their primary income continue to suffer irreparable harm. Also, to date, MindGeek and Pornhub have not paid settlements on cases arising under the new legislation. Most cases against internet providers have not survived a motion to dismiss. These civil actions also fail to address harm to victims outside the jurisdiction of countries with similar measures. If the goal is to bring an end to exploitation for profit on the internet, not merely to legislate morality and end sex-work in general, a more comprehensive and targeted solution is needed.
This article contemplates a corporate governance solution that could aid advances in technology by placing a limit on the reliance by company management on corporate structure and contractual relationships to disclaim responsibility and justify inaction. In a prior work, Corporate Family Matters, I propose a definition and governance regime for a particular type of corporate group—the corporate family. A corporate family is an enterprise formed by weaving corporations, partnerships, and LLCs together in a mix of public and private entities acting for the benefit of a parent corporation or for the personal gain of one or more leaders of the enterprise. Using MindGeek as an example, this Essay applies this definition to the enterprise, and explains how acknowledging the influence of MindGeek and treating the enterprise as a family can provide relief to victims while minimizing collateral harms
Bayesian Variational Inference in Keyword Identification and Multiple Instance Classification
Get PDFThis dissertation investigates (1) Variational Bayesian Semi-supervised Keyword Extraction and (2) Variational Bayesian Multimodal Multiple Instance Classification.
The expansion of textual data, stemming from various sources such as online product reviews and scholarly publications on scientific discoveries, has created a demand for the extraction of succinct yet comprehensive information. As a result, in recent years, efforts have been spent in developing novel methodologies for keyword extraction. Although many methods have been proposed to automatically extract keywords in the contexts of both unsupervised and fully supervised learning, how to effectively use partially observed keywords, such as author-specified keywords, remains an under-explored area. In Chapter 1, we propose a novel variational Bayesian semi-supervised (VBSS) keyword extraction approach, built on a recent Bayesian semi-supervised (BSS) technique that uses the information from a small set of known keywords to identify previously undetected ones. Our proposed VBSS method greatly enhances the computational efficiency of BSS via mean-field variational inference, coupled with data augmentation, which brings closed-form solutions at each step of the optimization process. Further, our numerical results show that VBSS offers enhanced accuracy for long texts and improved control over false discovery rates when compared with a list of state-of-the-art keyword extraction methods.
In Chapter 2, we apply mean-field variational inference on multiple instance learning (MIL). In MIL, objects are represented by bags of instances. Each instance shares the same feature set but has unique feature values. MIL aims to train models that predict bag-level outcomes based on these instances, making it a weakly supervised approach due to the lack of instance-level labels. While MIL methods focusing on binary classification are abundant, they often cannot identify which specific instances drive bag labels and have limited or little interpretability. Xiong et al. (2024) introduced MICProB, a Bayesian multiple instance classification (MIC) algorithm that addresses these issues. However, MICProB is computationally intensive and best suited for unimodal instances. To overcome these limitations, we propose a novel variational Bayesian multimodal MIC (vMMIC) algorithm. vMMIC handles diverse instance types and significantly improves computational efficiency through Bayesian variational inference, combined with data augmentation. We benchmark vMMIC against MICProB and many other MIC approaches on both simulated and real-world data. Results demonstrate vMMIC\u27s superior performance, computational efficiency, and interpretability
Where A-B Testing Goes Wrong: How Divergent Delivery Affects What Online Experiments Cannot (and Can) Tell You About How Customers Respond to Advertising
No full textMarketers use online advertising platforms to compare user responses to different ad content. However, platforms’ experimentation tools deliver ads to distinct, optimized, undetectable mixes of users that vary across ads, even during the test. As a result, the estimated - comparison from the data reflects the combination of ad content and algorithmic selection of users, which is different than what would have occurred under random exposure. We empirically demonstrate this “divergent delivery” pattern using data from an - test that we ran on a major ad platform. This paper explains how algorithmic targeting, user heterogeneity, and data aggregation conspire to confound the magnitude, and even the sign, of ad - test results, and what the implications are for different roles in the marketing organization with varying experimentation goals. We also consider the counterfactual case of disabling divergent delivery, where user types are balanced across ads. By extending the potential outcomes model of causal inference, we treat random assignment of ads and user exposure to ads as independent decisions. Since not all marketers have the same decision-making goals for these ad - tests, we offer prescriptive guidance to experimenters based on their needs
Leveraging Transformer Models for Genre Classification
Get PDFAs the digital music landscape continues to expand, the need for effective methods to understand and contextualize the diverse genres of lyrical content becomes increasingly critical. This research focuses on the application of transformer models in the domain of music analysis, specifically in the task of lyric genre classification. By leveraging the advanced capabilities of transformer architectures, this project aims to capture intricate linguistic nuances within song lyrics, thereby enhancing the accuracy and efficiency of genre classification. The relevance of this project lies in its potential to contribute to the development of automated systems for music recommendation and genre-based playlist creation. Moreover, understanding the linguistic features that define distinct musical genres through transformer-based models offers valuable insights into the underlying patterns and characteristics of lyrical content. The final pre-trained transformer model chosen for the final model is called DistilBERT, which is a “distilled” version of the popular pre-trained transformer model called BERT (Biodirectional Encoder Representations from Transformers). The implications, challenges, ethical concerns, and future research are discussed and are sought to be addressed
Honor Systems and Open World Activities in Red Dead Redemption II: Impact on Players with Symptoms of Anxiety and Depression
No full textThis research delves into the profound impact of honor systems and open-world activities on players\u27 mental well-being within the immersive world of Red Dead Redemption II (2018). The study uncovers intricate connections between gameplay choices, playstyles, and psychological needs satisfaction using self-determination theory (SDT), shedding light on the well-being effects for players experiencing phases of low mood. Operationalized measures were employed to assess pre-game self-reported anxiety and depression symptoms, engagement within honor and open-world activities, and post-engagement mental well-being. Data was sourced through an online questionnaire (n = 210) and meticulously analyzed to unveil insights into the engagement and playstyles adopted to alleviate players’ symptoms of anxiety and depression. The study\u27s findings contribute valuable knowledge to the expanding research on video games and their impact on mental health, advocating for game development that harnesses these interactive media’s positive influence on players\u27 well-being
The Case Law Of The Court Of Justice Of The EU On Art. 17 Of The 1999 Montreal Convention: An Evaluation From A Comparative Perspective
Get PDFThis paper analyzes the case law of the Court of Justice of the European Union (CJEU) on Article 17(1) of the 1999 Montreal Convention (MC99) regarding the liability of international air carriers for death or bodily injury to passengers. The interpretational principles and methods applied by the CJEU are examined, accounting also for the particularities of the EU legal order. Furthermore, the results reached by the CJEU are compared with the case law of other jurisdictions, mainly the US, and doctrinal writings. Nonetheless, this paper does not explore the pertinent issues from a de lege ferenda perspective.
The paper concludes that the judgments of the CJEU on Art. 17(1) MC99 have interpreted the notions of “passenger,” “accident,” and “bodily injury” broadly, in a passenger-friendly way. Although the interpretation of ‘passenger’ does not differ from the established case law in other jurisdictions, some aspects of the interpretation of “accident” and the interpretation of “bodily injury” significantly depart from the view currently prevailing among courts internationally. The CJEU has yet to rule on the scope of the exclusivity of the MC99, under Art. 29 thereof, regarding personal injury of passengers. However, the expansive interpretations of “accident” and “bodily injury” by the CJEU limit the practical effect of Article 29 compared to other jurisdictions. Given the regulatory influence that the EU exercises world- wide, the CJEU judgments might guide courts also outside the EU. Although this would bolster passenger protection, it would exacerbate the already fragmented application of the MC99 internationally