arXiv subjects

Ho-Chun Herbert Chang

Publications and source records attributed to Ho-Chun Herbert Chang.

25 records · Page 2Linked to original sources

arXiv2021

Targeted Ads and/as Racial Discrimination: Exploring Trends in New York City Ads for College Scholarships

This paper uses and recycles data from a third-party digital marketing firm, to explore how targeted ads contribute to larger systems of racial discrimination. Focusing on a case study of targeted ads for educational searches in New York City, it discusses data visualizations and mappings of trends in the advertisements' targeted populations alongside U.S census data corresponding to these target zipcodes. We summarize and reflect on the results to consider how internet platforms systemically and differentially target advertising messages to users based on race; the tangible harms and risks that result from an internet traffic system designed to discriminate; and finally, novel approaches and frameworks for further auditing systems amid opaque, black-boxed processes forestalling transparency and accountability.

cs.CY↗

arXiv2021

Social Bots and Social Media Manipulation in 2020: The Year in Review

The year 2020 will be remembered for two events of global significance: the COVID-19 pandemic and 2020 U.S. Presidential Election. In this chapter, we summarize recent studies using large public Twitter data sets on these issues. We have three primary objectives. First, we delineate epistemological and practical considerations when combining the traditions of computational research and social science research. A sensible balance should be struck when the stakes are high between advancing social theory and concrete, timely reporting of ongoing events. We additionally comment on the computational challenges of gleaning insight from large amounts of social media data. Second, we characterize the role of social bots in social media manipulation around the discourse on the COVID-19 pandemic and 2020 U.S. Presidential Election. Third, we compare results from 2020 to prior years to note that, although bot accounts still contribute to the emergence of echo-chambers, there is a transition from state-sponsored campaigns to domestically emergent sources of distortion. Furthermore, issues of public health can be confounded by political orientation, especially from localized communities of actors who spread misinformation. We conclude that automation and social media manipulation pose issues to a healthy and democratic discourse, precisely because they distort representation of pluralism within the public sphere.

cs.SI↗

arXiv2021

Tracking e-cigarette warning label compliance on Instagram with deep learning

The U.S. Food & Drug Administration (FDA) requires that e-cigarette advertisements include a prominent warning label that reminds consumers that nicotine is addictive. However, the high volume of vaping-related posts on social media makes compliance auditing expensive and time-consuming, suggesting that an automated, scalable method is needed. We sought to develop and evaluate a deep learning system designed to automatically determine if an Instagram post promotes vaping, and if so, if an FDA-compliant warning label was included or if a non-compliant warning label was visible in the image. We compiled and labeled a dataset of 4,363 Instagram images, of which 44% were vaping-related, 3% contained FDA-compliant warning labels, and 4% contained non-compliant labels. Using a 20% test set for evaluation, we tested multiple neural network variations: image processing backbone model (Inceptionv3, ResNet50, EfficientNet), data augmentation, progressive layer unfreezing, output bias initialization designed for class imbalance, and multitask learning. Our final model achieved an area under the curve (AUC) and [accuracy] of 0.97 [92%] on vaping classification, 0.99 [99%] on FDA-compliant warning labels, and 0.94 [97%] on non-compliant warning labels. We conclude that deep learning models can effectively identify vaping posts on Instagram and track compliance with FDA warning label requirements.

cs.SI↗

arXiv2020

Multi-Issue Bargaining With Deep Reinforcement Learning

Negotiation is a process where agents aim to work through disputes and maximize their surplus. As the use of deep reinforcement learning in bargaining games is unexplored, this paper evaluates its ability to exploit, adapt, and cooperate to produce fair outcomes. Two actor-critic networks were trained for the bidding and acceptance strategy, against time-based agents, behavior-based agents, and through self-play. Gameplay against these agents reveals three key findings. 1) Neural agents learn to exploit time-based agents, achieving clear transitions in decision preference values. The Cauchy distribution emerges as suitable for sampling offers, due to its peaky center and heavy tails. The kurtosis and variance sensitivity of the probability distributions used for continuous control produce trade-offs in exploration and exploitation. 2) Neural agents demonstrate adaptive behavior against different combinations of concession, discount factors, and behavior-based strategies. 3) Most importantly, neural agents learn to cooperate with other behavior-based agents, in certain cases utilizing non-credible threats to force fairer results. This bears similarities with reputation-based strategies in the evolutionary dynamics, and departs from equilibria in classical game theory.

cs.MA↗

arXiv2020

Elitism in Mathematics and Inequality

The Fields Medal, often referred as the Nobel Prize of mathematics, is awarded to no more than four mathematician under the age of 40, every four years. In recent years, its conferral has come under scrutiny of math historians, for rewarding the existing elite rather than its original goal of elevating mathematicians from under-represented communities. Prior studies of elitism focus on citational practices and sub-fields; the structural forces that prevent equitable access remain unclear. Here we show the flow of elite mathematicians between countries and lingo-ethnic identity, using network analysis and natural language processing on 240,000 mathematicians and their advisor-advisee relationships. We found that the Fields Medal helped integrate Japan after WWII, through analysis of the elite circle formed around Fields Medalists. Arabic, African, and East Asian identities remain under-represented at the elite level. Through analysis of inflow and outflow, we rebuts the myth that minority communities create their own barriers to entry. Our results demonstrate concerted efforts by international academic committees, such as prize-giving, are a powerful force to give equal access. We anticipate our methodology of academic genealogical analysis can serve as a useful diagnostic for equality within academic fields.

math.HO↗

arXiv2019

Co-Contagion Diffusion on Multilayer Networks

This study examines the interface of three elements during co-contagion diffusion: the \textbf{synergy} between contagions, the \textbf{dormancy} rate of each individual contagion, and the \textbf{multiplex network topology}. Dormancy is defined as a weaker form of "immunity," where dormant nodes no longer actively participate in diffusion, but are still susceptible to infection. The proposed model extends the literature on threshold models, and demonstrates intricate interdependencies between different graph structures. Our simulations show that first, the faster contagion induces branching on the slower contagion; second, shorter characteristic path lengths diminish the impact of dormancy in lowering diffusion. Third, when two long-range graphs are paired, the faster contagion depends on both dormancy rates, whereas the slower contagion depends only on its own; fourth, synergistic contagions are less sensitive to dormancy, and have a wider window to diffuse. Furthermore, when long-range and spatially constrained graphs are paired, ring vaccination occurs on the spatial graph and produces partial diffusion, due to dormant, surrounding nodes. The spatial contagion depends on both dormancy rates whereas the long-range contagion depends on only its own.

cs.SI↗

arXiv2018

Co-Diffusion of Social Contagions

Prior social contagion models consider the spread of either one contagion at a time on interdependent networks or multiple contagions on single layer networks or under assumptions of competition. We propose a new threshold model for the diffusion of multiple contagions. Individuals are placed on a multiplex network with a periodic lattice and a random-regular-graph layer. On these population structures, we study the interface between two key aspects of the diffusion process: the level of synergy between two contagions, and the rate at which individuals become dormant after adoption. Dormancy is defined as a looser form of immunity that models the ability to spread without resistance. Monte Carlo simulations reveal lower synergy makes contagions more susceptible to percolation, especially those that diffuse on lattices. Faster diffusion of one contagion with dormancy probabilistically blocks the diffusion of the other, in a way similar to ring vaccination. We show that within a band of synergy, contagions on the lattices undergo bimodal or trimodal branching if they are the slower diffusing contagion.

cs.SI↗