Skip to main content

News & Events

Events

 

RSS

November 19, 2018, 12:00 PM
Professor Ran Canetti: From Soft Classifers to Hard Decisions: How Fair Can We Be?

[Monday, November 19, 2018 at 12:00 PM in Wegmans Hall 1400] As society is charging ahead with the prevalent use of (often opaque) algorithms in decision-making processes regarding people, it has become critical to study the societal implications of such algorithms. This study is fascinating, in that it forces us to mathematically formulate and quantify concepts such as fairness and social justice, which have traditionally been thought of in qualitative and subjective terms. In particular, this study is teasing out inherent inconsistencies in traditional thinking and helping us clarify our social goals. We consider one slice of this study: The need to balance the errors across societal groups in algorithmic decision-making based on imperfect information. Here some widely used statistical measures error-balance turn out to be contradictory (Kleinberg et al. '16, Chouldechova '17). We propose a way to - at least partially - evade this bound. Specifically, we observe that if the algorithm is allowed to `defer' on some decisions (that is, the algorithm can avoid making some decisions and hand them off to a separate process which is presumably more expensive and more precise), then the errors in the non-deferred decisions will be balanced across the considered societal groups by all statistical measures currently under consideration. We further demonstrate concrete ways for using deferrals in the natural context of post-processing ``soft’’ classifiers that output a numerical score into a binary decision. We evaluate our algorithms using the COMPAS data set from 2016 and consider potential societal implications of deferrals.


November 21, 2018, 11:00 AM
Solomon Abiola: Node: A Machine Learning Approach to Disease Detection in Wearable and Mobile Networks

[Wednesday, November 21, 2018 at 11:00 AM in Wegmans Hall 2506] This survey paper presents both a perspective on the current state of the field of machine learning in healthcare as it pertains to infectious disease modeling using network science and data mining techniques. While offering an overview of ongoing original research and early conclusions from such findings using data obtained from the Ebola outbreak in Lagos, Nigeria, 2015. In 2015 the “Node” application developed in 2013 at Princeton University was deployed in Lagos, Nigeria at the Lagos University Teaching Hospital, Nigerian Institute of Medical Research, and the University of Lagos. The app consisted of a real-time geospatial location feature (resolution less than 10 meters) and a WHO quality of life survey. For three months, August – October 2015 at 5-minute intervals location information was collected. Our network analysis of human mobility patterns indicated that in Lagos, Nigeria and potentially other megacities early weekdays present the most likely day on which an outbreak may occur. Secondly, should such an outbreak be likely public health officials and providers with limited resources should take prudent action to secure individuals with high centrality in the network. Large-scale implementation given the successful nature of validating the application would allow these interventions to be studied against ongoing diseases such as malaria or tuberculosis.


November 26, 2018, 12:00 PM
Rob Shearer: The Promise of AR and New Challenges for Silicon

[Monday, November 26, 2018 at 12:00 PM in Wegmans Hall 1400] A look at how AR & VR are evolving and some possible applications and use cases. Then we’ll review some of the key technology challenges that need to be overcome before widespread deployment of AR & VR will happen and the breakthroughs in silicon that are required to enable this future. We need substantial advancements beyond the state of the art to meet the silicon performance/power targets.


November 26, 2018, 01:00 PM
Brian Dickinson: Leveraging Worldwide Population Level Movement Data

[Monday, November 26, 2018 at 1:00 PM in Wegmans Hall 2506]

Many questions about the movement of populations have recently become tractable at unprecedented scale. In the past we have relied primarily on national censuses, community surveys, commuter surveys, and similar measures to estimate population movements. These measures, however, are limited in their sample sizes. They also tend to vary between countries and are nearly entirely absent in the developing world. At the turn of the century this began to change as logs from cell tower triangulation of individual mobile devices became available for study. Still, these new datasets had many of the same geographic limitations being confined primarily to carriers in parts of Europe. With the advent of smartphones a large percentage of the world's population now carry GPS enabled computers with them everywhere which collect fine grained logs of movement. Additionally, since the market is dominated by Apple's iOS and Google's Android devices, this data is nearly uniform worldwide, eliminating regional differences in collection methods. This data provides an unprecedented insight into population level movement patterns.

In this proposal we will consider what we can learn both from patterns of movement alone and from correlations between location, movement, and a variety of outside factors. For this analysis we make use of Google's Location History data; this is done in the least invasive way possible using the latest differential privacy techniques to protect the privacy of the millions of individuals whose data is aggregated for our population level analysis. We will look at the hierarchical structure of mobility in urban areas worldwide and the connection of this structure to public health and environmental factors. We will explore the connections between movement patterns and socioeconomic indicators to better understand how different factors affect individual's activities and how we can better design our cities and infrastructure to benefit everyone. Finally, we will consider how population mobility may replace nighttime satellite imagery as a global metric for predicting the extent of human development and economic output.

In the future we plan to explore applications in epidemiology modeling the spread of influenza using real-time information on movement patterns within and between cities worldwide. We also hope to identify geographic differences across a range of currently relevant factors such as perception and dissemination of uncivil speech. We include a proposed timeline for these studies in the final section of this proposal.

Advisor: Professor Henry Kautz (Computer Science)

Committee Members: Professor Jiebo Luo (Computer Science), Professor M. Ehsan Hoque (Computer Science), Professor Gourab Ghoshal (Physics)


November 26, 2018, 02:30 PM
Anis Zaman: Combining traditional and non-traditional data stream for understanding mental health

[Monday, November 26, 2018 at 2:30 PM in Wegmans Hall 2506]

Mental health can be conceptualized as a complex, emergent phenomenon resulting from the interaction between biological, psychological, social, environmental, and other contextual factors. Traditionally, care providers assess one's mental health when an individual consults a medical or behavioral health professional. Different events in our quotidian life trigger various emotional and cognitive responses. Since much of our lives is now lived online, online activities on digital platforms may provide valuable insights and understandings of an individual’s mental states. Recently, social computer scientists have attempted to model mental health phenomena based on data derived from social media activities. We propose a framework to incorporate online activities with longitudinal ground truth data about an individual for detecting various mental health phenomena. We hypothesize that both the routine experiences of daily living, as well as specific, significant life events, invoke particular patterns that can be detected in online activities. A close monitoring of these experiences and events, along with search activity and other online behavior, can inform systems to provide early diagnosis and detection of depression and related mental health conditions.

Advisor: Professor Henry Kautz (Computer Science)

Committee: Professor Jiebo Luo (Computer Science), Professor M. Ehsan Hoque (Computer Science), Professor Vincent Silenzio (URMC/Psychiatry)


November 26, 2018, 04:00 PM
Nabil Hossain: Exploring Creativity and Style in Language Generation

[Monday, November 26, 2018 at 4:00 PM in Wegmans Hall 2506]

Building systems for understanding creative language is a notoriously difficult problem in AI. The task not only demands machine learning algorithms that incorporate world knowledge, reasoning, and perception but also requires substantial effort in annotating training data. In this proposal, we explore the generation and detection of creative language through three areas where creativity is involved: general and topical (news based) humor generation; creative slang in politically-oriented online speech; and style transfer to deal with issues of political bias in headlines and hate in comments posted in online communities.

In the first area, we develop datasets and algorithms for humor generation (i) in Mad Libs®, where the goal is to fill-in the blanks in a short story to make it funny, and (ii) in news headlines, where our objective is to achieve humor by applying small edits to original headlines. In the second area, we investigate the generation and usage of creative slang in online political discourse among reader communities within and across major news sources. This domain provides an interesting intersection of humor and hate speech, because many slang terms are both hateful and intended to be humorous. In the third area, we explore various deep-learning approaches for style transfer, applying these methods to re-write creative language. We specifically investigate the use of style transfer to reduce political bias in news headlines and to re-write uncivil online comments to make them civil and more acceptable to a general audience.

Advisor: Professor Henry Kautz (Computer Science)

Committee: Professor Lenhart Schubert (Computer Science), Professor Chenliang Xu (Computer Science), Professor Aaron White (Linguistics), Professor Curtis Signorino (Political Science), John Krumm (Microsoft)


November 30, 2018, 01:00 PM
Guoyu Lu: TBD

[Friday, November 30, 2018 at 1:00 PM in Wegmans Hall 2506] TBD


December 3, 2018, 12:00 PM
Professor Dongyoon Lee: Fuzzing and Securing the Server-Side Event-Driven Architecture

[Monday, December 03, 2018 at 12:00 PM in Wegmans Hall 1400] The software development community is adopting the Event-Driven Architecture (EDA) to provide scalable web services, most prominently through Node.js. Though the EDA scales well, it comes with two inherent risks: concurrency errors and Event Handler Poisoning (EHP) Denial of Service attacks. Just as thread-based programs can have concurrency errors between unordered threads, event-driven programs may have them between unordered events. When an EDA-based server multiplexes many clients onto few threads, a blocked thread (EHP) renders the whole server unresponsive. In this talk, I present Node.fz and Node.cure to address these problems. First, Node.fz provides a schedule fuzzing test tool that randomly perturbs the execution of a Node.js program, allowing Node.js developers to explore a variety of possible schedules during testing. Second, Node.cure proposes First-Class Timeouts, which incorporates timeouts at the EDA framework level, defending Node.js applications against all known EHP attacks.


December 4, 2018, 10:00 AM
Professor Yelin Kim: Multimodal Emotion Recognition: Quantifying Dynamics andStructure in Audio-Visual Expressive Speech

[Tuesday, December 04, 2018 at 10:00 AM in Wegmans Hall 2506] The rise of AI assistant systems, including Google Home, Apple Siri, and Amazon Echo, brings the urgent need for increased and deeper understanding of users. In this talk, I will present algorithmic and statistical methods for analyzing audio-visual human behavior, particularly focusing on emotional and social signals inferred from speech and facial expressions. These methods can provide emotional intelligence to AI systems. However, developing automatic emotion recognition systems is challenging since emotional expressions are complex, dynamic, inherently multimodal, and are entangled with other factors of modulation (e.g. speech generation and emphasis). I will present several algorithms to address these fundamental challenges in emotion recognition: (i) cross-modal modeling methods that capture and control for interactions between individual facial regions and speech using the Minimum Description Length (MDL) principle-based segmentation; (ii) localization and prediction of events with salient emotional behaviors using a max-margin optimization and dynamic programming; and (iii) temporal modeling methods to learn co-occurrence patterns between emotional behaviors and emotion label noise. These algorithms have enabled advancements in the modeling of audio-visual emotion recognition systems and increased the understanding of the underlying dynamic and multimodal structure of affective communication (e.g., cross-modal interaction, temporal structure, and inherent perceptual ambiguity).


December 10, 2018, 12:00 PM
Professor Zhong Shao: CertiKOS: From Hacker-Resistant OS to Certified Heterogeneous Systems

[Monday, December 10, 2018 at 12:00 PM in Wegmans Hall 1400] Computer science is being transformed by new exciting "vertical" research themes such as IoTs, self-driving cars, blockchains, and robotics. These new themes often require complex computing platforms involving not only heterogeneous hardware and software components but also physical, human, and social entities. The successful deployment of these technologies would critically depend on our ability to establish strong properties such as safety, security, resilience, resource efficiency, fairness, and privacy. In this talk, I present a novel compositional approach for building certified heterogeneous systems that aim to provide many such trustworthiness guarantees. We advocate abstraction over a particularly rich class of specification and present new methodologies and tools for formally specifying, programming, verifying, and composing abstraction layers. Using these new technologies, we have successfully developed the CertiKOS certified OS kernel and verified its contextual functional correctness in the Coq proof assistant. This is the world's first proof of functional correctness of a complete, general-purpose concurrent OS kernel with fine-grained locking. We show how to extend our base kernel with new features such as virtualization, interrupts and device drivers, and end-to-end information flow security, and how to quickly adapt existing verified layers to build new certified kernels for modern heterogeneous platforms.


January 18, 2019, 01:00 PM
Yu Kong: TBD

[Friday, January 18, 2019 at 1:00 PM in Wegmans Hall 2506] TBD