Podcasts and Videos
Plan to join us WEDS, 09 Dec for a very, VERY special TWed as the Tetherless World Constellation holds another "virtual" version of our end-of-term Graduate Research
"Lightning Talks." TWed Lightning Talks are a great way for the TWC community and friends to learn of the wide range of amazing research happening in the Tetherless World, and "a good time is had by all!"
Lightning talks are VERY short --- approx. 2-3 minute! --- summaries by our students of current research work, with no NO SLIDES and only brief "crib notes."
This talk will be an introduction to the study of privacy attacks in the context of machine learning, aimed at those unfamiliar with the literature. We will discuss who the stakeholders are, what information may be attacked, how it may be attacked, and why. The “how” will be at a high level, illustrated through some specific examples of privacy attacks. Much of the material will be from a recent survey of privacy attacks by Maria Rigaki and Sebastian Garcia at Czech Technical University in Prague, although their threat model will be extended slightly to consider cases that include synthetic data. The goal of the talk is to give the audience an appreciation of some of the complications of privacy preservation (i.e. that it’s not as simple as it may be assumed to be) and familiarity with some of the terminology.
Miao will discuss the tools and techniques used to assess, visualize, and improve equity in clinical trials. A set of novel equity metrics for clinical trials is constructed from Machine Learning (ML) Fairness Research to quantify inequities of various subgroups defined over multiple demographic or clinical characteristics, such as Hispanic female subjects who are underweight or no-Hispanic black male subjects aged over 64 and with high fasting glucose level. A tool called TrialEquity, which is developed based on the proposed equity metrics, is designed to provide insights to improve the clinical trial equity and health equity, with specific considerations for diverse user groups including clinicians, researchers, and health policy advocates. The tool is able to design new equitable clinical trials, evaluate ongoing/conducted studies, provide remedial advice for inequitable trials, accommodate how evidence from trials applies to the individual needs of patients, and guide equitable decisions for users.
Challenge: Midlife mortality rates are rising in the United States (US), while in many other nations, mortality rates are decreasing. For example, Stein et al. (2017) found that “Deaths of Despair” due to suicide and substance abuse have increased dramatically among white males between the ages of 25-64 particularly in rural America. The MortalityMinder (MM) app’s goal is to enable healthcare researchers, providers, payers, and policy makers to gain actionable insights into how, where, and why midlife mortality rates are rising in the US.System
Description and Purpose: Using county-level data on mortality rates from CDC WONDER, MM explores mortality trends for adults ages 25-64 in the US from 2000 to 2017. Using county-level surveillance data from County Health Rankings, MM identifies social and economic factors associated with mortality trends at the county level for the US and individual states. The user selects the region (specific state or US) and the cause of death (All Causes,Cancer, Cardiovascular, or Deaths of Despair). MM divides counties into mortality risk groups using clustering and then finds statistical associations between groups and putative risk factors. MM dynamically creates three analysis and visualization infographics, each addressing a different question:
Mosquitoes are responsible for transfer of many vector-borne diseases. Dengue is one such viral infection that is transmitted by the Aedes mosquito. It is preventable but still the number of Dengue cases have risen 30-fold in the past 50 years. In several countries in south American continent and Asia, dengue is one of the leading causes of death. It is mainly found in tropical and sub-tropical regions, particularly surrounding urban and semi-urban areas.
Historically, there has been an intensive increase in the number of dengue cases from 2000-2010 and, if adequately explored, essential information can be retrieved. Thus, we decided to develop Dengue Spread Information System (DSIS), a geographic-health information system designed to highlight the spread of dengue cases in Iquitos, Peru, and San Juan, Puerto Rico from 1990 to 2013. The application is aimed at citizens, travelers, policymakers and researchers to analyze and interpret the change in risk factors leading to dengue outbreaks and develop essential early warning applications and policies to counter future dengue outbreaks.
The application portrays an interactive map for the two cities with additional information about temperature and humidity. The application is accompanied by exploratory data analysis on several risk factors impacting Dengue spread which can aid and facilitate research in the domain so that Dengue spread can be prevented.
DESCRIPTION: The goals of commonsense reasoning systems include being able to answer commonsense reasoning questions. In order to compare systems, a number of benchmark question sets have arisen. Leaderboards have emerged to act as hubs for hosting benchmarks and supporting infrastructure that accepts submissions of commonsense reasoning systems that then get scored against the benchmarks. These benchmarks vary in structure. Some provide questions and answer choices, while others may provide factual observations and require reasoners to choose the most appropriate hypothesis to explain them.
Recently, there is an increasing effort to incorporate structured knowledge in these systems, largely based on machine-learning techniques, as a way to improve their overall score against benchmarks. In this talk, we will present and discuss our current efforts in supporting this goal, in the context of the Machine Commonsense Project. It includes a Benchmark Ontology, which provides a common vocabulary to allow diverse benchmarks to be compared, integrated, and to support the analysis of systems and machine-learning language models. This talk will discuss its design decisions and showcases how it is currently supporting the development of a Benchmark tool.