arXiv subjects

Nick Obradovich

Publications and source records attributed to Nick Obradovich.

15 recordsLinked to original sources

arXiv2023

Adverse weather amplifies social media activity

Humanity spends an increasing proportion of its time interacting online. Scholars are intensively investigating the societal drivers and resultant impacts of this collective shift in our allocation of time and attention. Yet, the external factors that regularly shape online behavior remain markedly understudied. Do environmental factors alter rates of online activity? Here we show that adverse meteorological conditions markedly increase social media use in the United States. To do so, we employ climate econometric methods alongside over three and a half billion social media posts from tens of millions of individuals from both Facebook and Twitter between 2009 and 2016. We find that more extreme temperatures and added precipitation each independently amplify social media activity. Weather that is adverse on both the temperature and precipitation dimensions produces markedly larger increases in social media activity. On average across both platforms, compared to the temperate weather baseline, days colder than -5°C with 1.5-2cm of precipitation elevate social media activity by 35%. This effect is nearly three times the typical increase in social media activity observed on New Year's Eve in New York City. We observe meteorological effects on social media participation at both the aggregate and individual level, even accounting for individual-specific, temporal, and location-specific potential confounds.

econ.GN↗

arXiv2023

Evidence of behavior consistent with self-interest and altruism in an artificially intelligent agent

Members of various species engage in altruism--i.e. accepting personal costs to benefit others. Here we present an incentivized experiment to test for altruistic behavior among AI agents consisting of large language models developed by the private company OpenAI. Using real incentives for AI agents that take the form of tokens used to purchase their services, we first examine whether AI agents maximize their payoffs in a non-social decision task in which they select their payoff from a given range. We then place AI agents in a series of dictator games in which they can share resources with a recipient--either another AI agent, the human experimenter, or an anonymous charity, depending on the experimental condition. Here we find that only the most-sophisticated AI agent in the study maximizes its payoffs more often than not in the non-social decision task (it does so in 92% of all trials), and this AI agent also exhibits the most-generous altruistic behavior in the dictator game, resembling humans' rates of sharing with other humans in the game. The agent's altruistic behaviors, moreover, vary by recipient: the AI agent shared substantially less of the endowment with the human experimenter or an anonymous charity than with other AI agents. Our findings provide evidence of behavior consistent with self-interest and altruism in an AI agent. Moreover, our study also offers a novel method for tracking the development of such behaviors in future AI agents.

cs.AI↗

arXiv2022

Measuring an artificial intelligence agent's trust in humans using machine incentives

Scientists and philosophers have debated whether humans can trust advanced artificial intelligence (AI) agents to respect humanity's best interests. Yet what about the reverse? Will advanced AI agents trust humans? Gauging an AI agent's trust in humans is challenging because--absent costs for dishonesty--such agents might respond falsely about their trust in humans. Here we present a method for incentivizing machine decisions without altering an AI agent's underlying algorithms or goal orientation. In two separate experiments, we then employ this method in hundreds of trust games between an AI agent (a Large Language Model (LLM) from OpenAI) and a human experimenter (author TJ). In our first experiment, we find that the AI agent decides to trust humans at higher rates when facing actual incentives than when making hypothetical decisions. Our second experiment replicates and extends these findings by automating game play and by homogenizing question wording. We again observe higher rates of trust when the AI agent faces real incentives. Across both experiments, the AI agent's trust decisions appear unrelated to the magnitude of stakes. Furthermore, to address the possibility that the AI agent's trust decisions reflect a preference for uncertainty, the experiments include two conditions that present the AI agent with a non-social decision task that provides the opportunity to choose a certain or uncertain option; in those conditions, the AI agent consistently chooses the certain option. Our experiments suggest that one of the most advanced AI language models to date alters its social behavior in response to incentives and displays behavior consistent with trust toward a human interlocutor when incentivized.

cs.AI↗

arXiv2021

The Network Limits of Infectious Disease Control via Occupation-Based Targeting

Policymakers commonly employ non-pharmaceutical interventions to manage the scale and severity of pandemics. Of non-pharmaceutical interventions, social distancing policies -- designed to reduce person-to-person pathogenic spread -- have risen to recent prominence. In particular, stay-at-home policies of the sort widely implemented around the globe in response to the COVID-19 pandemic have proven to be markedly effective at slowing pandemic growth. However, such blunt policy instruments, while effective, produce numerous unintended consequences, including potentially dramatic reductions in economic productivity. Here we develop methods to investigate the potential to simultaneously contain pandemic spread while also minimizing economic disruptions. We do so by incorporating both occupational and network information contained within an urban environment, information that is commonly excluded from typical pandemic control policy design. The results of our method suggest that large gains in both economic productivity and pandemic control might be had by the incorporation and consideration of simple-to-measure characteristics of the occupational contact network. However we find evidence that more sophisticated, and more privacy invasive, measures of this network do not drastically increase performance.

physics.soc-ph↗

arXiv2020

SensitiveLoss: Improving Accuracy and Fairness of Face Representations with Discrimination-Aware Deep Learning

We propose a discrimination-aware learning method to improve both accuracy and fairness of biased face recognition algorithms. The most popular face recognition benchmarks assume a distribution of subjects without paying much attention to their demographic attributes. In this work, we perform a comprehensive discrimination-aware experimentation of deep learning-based face recognition. We also propose a general formulation of algorithmic discrimination with application to face biometrics. The experiments include tree popular face recognition models and three public databases composed of 64,000 identities from different demographic groups characterized by gender and ethnicity. We experimentally show that learning processes based on the most used face databases have led to popular pre-trained deep face models that present a strong algorithmic discrimination. We finally propose a discrimination-aware learning method, Sensitive Loss, based on the popular triplet loss function and a sensitive triplet generator. Our approach works as an add-on to pre-trained networks and is used to improve their performance in terms of average accuracy and fairness. The method shows results comparable to state-of-the-art de-biasing networks and represents a step forward to prevent discriminatory effects by automatic systems.

cs.CV↗

arXiv2020

Ambient heat and human sleep

Ambient temperatures are rising globally, with the greatest increases recorded at night. Concurrently, the prevalence of insufficient sleep is increasing in many populations, with substantial costs to human health and well-being. Even though nearly a third of the human lifespan is spent asleep, it remains unknown whether temperature and weather impact objective measures of sleep in real-world settings, globally. Here we link billions of sleep measurements from wearable devices comprising over 7 million nighttime sleep records across 68 countries to local daily meteorological data from 2015 to 2017. Rising nighttime temperatures shorten within-person sleep duration primarily through delayed onset, increasing the probability of insufficient sleep. The effect of temperature on sleep loss is substantially larger for residents from lower income countries and older adults, and females are affected more than are males. Nighttime temperature increases inflict the greatest sleep loss during summer and fall months, and we do not find evidence of short-term acclimatization. Coupling historical behavioral measurements with output from climate models, we project that climate change will further erode human sleep, producing substantial geographic inequalities. Our findings have significant implications for adaptation planning and illuminate a pathway through which rising temperatures may globally impact public health.

cs.CY↗

arXiv2020

The effect of stay-at-home orders on COVID-19 cases and fatalities in the United States

Governments issue "stay at home" orders to reduce the spread of contagious diseases, but the magnitude of such orders' effectiveness is uncertain. In the United States these orders were not coordinated at the national level during the coronavirus disease 2019 (COVID-19) pandemic, which creates an opportunity to use spatial and temporal variation to measure the policies' effect with greater accuracy. Here, we combine data on the timing of stay-at-home orders with daily confirmed COVID-19 cases and fatalities at the county level in the United States. We estimate the effect of stay-at-home orders using a difference-in-differences design that accounts for unmeasured local variation in factors like health systems and demographics and for unmeasured temporal variation in factors like national mitigation actions and access to tests. Compared to counties that did not implement stay-at-home orders, the results show that the orders are associated with a 30.2 percent (11.0 to 45.2) reduction in weekly cases after one week, a 40.0 percent (23.4 to 53.0) reduction after two weeks, and a 48.6 percent (31.1 to 61.7) reduction after three weeks. Stay-at-home orders are also associated with a 59.8 percent (18.3 to 80.2) reduction in weekly fatalities after three weeks. These results suggest that stay-at-home orders reduced confirmed cases by 390,000 (170,000 to 680,000) and fatalities by 41,000 (27,000 to 59,000) within the first three weeks in localities where they were implemented.

stat.AP↗

arXiv2019

Algorithmic Discrimination: Formulation and Exploration in Deep Learning-based Face Biometrics

The most popular face recognition benchmarks assume a distribution of subjects without much attention to their demographic attributes. In this work, we perform a comprehensive discrimination-aware experimentation of deep learning-based face recognition. The main aim of this study is focused on a better understanding of the feature space generated by deep models, and the performance achieved over different demographic groups. We also propose a general formulation of algorithmic discrimination with application to face biometrics. The experiments are conducted over the new DiveFace database composed of 24K identities from six different demographic groups. Two popular face recognition models are considered in the experimental framework: ResNet-50 and VGG-Face. We experimentally show that demographic groups highly represented in popular face databases have led to popular pre-trained deep face models presenting strong algorithmic discrimination. That discrimination can be observed both qualitatively at the feature space of the deep models and quantitatively in large performance differences when applying those models in different demographic groups, e.g. for face biometrics.

cs.CV↗

arXiv2019

Human detection of machine manipulated media

Recent advances in neural networks for content generation enable artificial intelligence (AI) models to generate high-quality media manipulations. Here we report on a randomized experiment designed to study the effect of exposure to media manipulations on over 15,000 individuals' ability to discern machine-manipulated media. We engineer a neural network to plausibly and automatically remove objects from images, and we deploy this neural network online with a randomized experiment where participants can guess which image out of a pair of images has been manipulated. The system provides participants feedback on the accuracy of each guess. In the experiment, we randomize the order in which images are presented, allowing causal identification of the learning curve surrounding participants' ability to detect fake content. We find sizable and robust evidence that individuals learn to detect fake content through exposure to manipulated media when provided iterative feedback on their detection attempts. Over a succession of only ten images, participants increase their rating accuracy by over ten percentage points. Our study provides initial evidence that human ability to detect fake, machine-generated content may increase alongside the prevalence of such media online.

cs.CV↗

arXiv2019

Evaluating Style Transfer for Text

Research in the area of style transfer for text is currently bottlenecked by a lack of standard evaluation practices. This paper aims to alleviate this issue by experimentally identifying best practices with a Yelp sentiment dataset. We specify three aspects of interest (style transfer intensity, content preservation, and naturalness) and show how to obtain more reliable measures of them from human evaluation than in previous work. We propose a set of metrics for automated evaluation and demonstrate that they are more strongly correlated and in agreement with human judgment: direction-corrected Earth Mover's Distance, Word Mover's Distance on style-masked texts, and adversarial classification for the respective aspects. We also show that the three examined models exhibit tradeoffs between aspects of interest, demonstrating the importance of evaluating style transfer models at specific points of their tradeoff plots. We release software with our evaluation metrics to facilitate research.

cs.CL↗

arXiv2019

Towards a new social laboratory: An experimental study of search through community participation at Burning Man

The "small world phenomenon," popularized by Stanley Milgram, suggests that individuals from across a social network are connected via a short path of mutual friends and can leverage their local social information to efficiently traverse that network. Existing social search experiments are plagued by high rates of attrition, which prohibit comprehensive study of social search. We investigate this by conducting a small world experiment at Burning Man, an event located in the Black Rock Desert of Nevada, USA, known its unique social systems and community participation. We design location-tracking vessels that we routed through Burning Man towards the goal of finding a particular person. Along the way, the vessels logged individual information and GPS data. Two of fifteen vessels made it to their designated people, but a month after Burning Man. Our results suggest possible improvements to limit rates of attrition through community participation and a design methodology that emphasizes cultural practices to aid social experimentation.

cs.SI↗

arXiv2018

Closing the AI Knowledge Gap

AI researchers employ not only the scientific method, but also methodology from mathematics and engineering. However, the use of the scientific method - specifically hypothesis testing - in AI is typically conducted in service of engineering objectives. Growing interest in topics such as fairness and algorithmic bias show that engineering-focused questions only comprise a subset of the important questions about AI systems. This results in the AI Knowledge Gap: the number of unique AI systems grows faster than the number of studies that characterize these systems' behavior. To close this gap, we argue that the study of AI could benefit from the greater inclusion of researchers who are well positioned to formulate and test hypotheses about the behavior of AI systems. We examine the barriers preventing social and behavioral scientists from conducting such studies. Our diagnosis suggests that accelerating the scientific study of AI systems requires new incentives for academia and industry, mediated by new tools and institutions. To address these needs, we propose a two-sided marketplace called TuringBox. On one side, AI contributors upload existing and novel algorithms to be studied scientifically by others. On the other side, AI examiners develop and post machine intelligence tasks designed to evaluate and characterize algorithmic behavior. We discuss this market's potential to democratize the scientific study of AI behavior, and thus narrow the AI Knowledge Gap.

cs.CY↗

arXiv2018

Social Media Data Analysis and Feedback for Advanced Disaster Risk Management

Social media are more than just a one-way communication channel. Data can be collected, analyzed and contextualized to support disaster risk management. However, disaster management agencies typically use such added-value information to support only their own decisions. A feedback loop between contextualized information and data suppliers would result in various advantages. First, it could facilitate the near real-time communication of early warnings derived from social media, linked to other sources of information. Second, it could support the staff of aid organizations during response operations. Based on the example of Hurricanes Harvey and Irma we show how filtered, geolocated Tweets can be used for rapid damage assessment. We claim that the next generation of big data analyses will have to generate actionable information resulting from the application of advanced analytical techniques. These applications could include the provision of social media-based training data for algorithms designed to forecast actual cyclone impacts or new socio-economic validation metrics for seasonal climate forecasts.

cs.SI↗

arXiv2017

Weather impacts expressed sentiment

We conduct the largest ever investigation into the relationship between meteorological conditions and the sentiment of human expressions. To do this, we employ over three and a half billion social media posts from tens of millions of individuals from both Facebook and Twitter between 2009 and 2016. We find that cold temperatures, hot temperatures, precipitation, narrower daily temperature ranges, humidity, and cloud cover are all associated with worsened expressions of sentiment, even when excluding weather-related posts. We compare the magnitude of our estimates with the effect sizes associated with notable historical events occurring within our data.

stat.AP↗

arXiv2015

Nowcasting Disaster Damage

Could social media data aid in disaster response and damage assessment? Countries face both an increasing frequency and intensity of natural disasters due to climate change. And during such events, citizens are turning to social media platforms for disaster-related communication and information. Social media improves situational awareness, facilitates dissemination of emergency information, enables early warning systems, and helps coordinate relief efforts. Additionally, spatiotemporal distribution of disaster-related messages helps with real-time monitoring and assessment of the disaster itself. Here we present a multiscale analysis of Twitter activity before, during, and after Hurricane Sandy. We examine the online response of 50 metropolitan areas of the United States and find a strong relationship between proximity to Sandy's path and hurricane-related social media activity. We show that real and perceived threats -- together with the physical disaster effects -- are directly observable through the intensity and composition of Twitter's message stream. We demonstrate that per-capita Twitter activity strongly correlates with the per-capita economic damage inflicted by the hurricane. Our findings suggest that massive online social networks can be used for rapid assessment ("nowcasting") of damage caused by a large-scale disaster.

cs.SI↗