2nd Graduate Research Symposium and Prospective Student Visit Day (2016)

W401 PBB audience and presenter during an afternoon symposium talkWelcome to the 2nd Iowa Computer Science Graduate Research Symposium (2016). Senior CS PhD students presented talks on their latest research, showcasing a variety of CS research areas including algorithms, mobile computing, networks, programming languages, and virtual reality. This was followed by a "keynote" by a CS faculty member. Talks were intended for a wide audience with interest in CS, including CS juniors and seniors.

Coinciding with the Graduate Research Symposium, the Computer Science (CS) department hosted a Prospective Student Visit Day for individuals from around the country interested in graduate programs in CS. We are looking for strong students with diverse backgrounds to join our MCS and PhD programs. A wide variety of research areas are represented by our world-class faculty including algorithms, computational epidemiology, distributed computing, human-computer interaction, machine learning, massive data algorithms and technology, mobile computing, networks, programming languages, text mining, and virtual reality. The talks presented by current CS graduate students at the Symposium are excellent examples of the exciting research taking place here in Iowa City!

Schedule

Friday, Nov 4, 2016

Morning Sessions (in MacLean Hall) For Visiting/Prospective Students

10-11am

Overview of Graduate Programs, Sriram Pemmaraju, Director of Graduate Studies

11-12noon

Visiting students will meet with Computer Science faculty members

12noon-1:30

Visiting students will meet and have lunch with Computer Science Graduate Students
 

Graduate Student Research Symposium (in W401 Pappajohn Business Building)

1:35-2:35pm

Session One

1:35pm

Baoluo Meng

Relational Constraints Solving via SMT

1:55pm

Shabih Hasan

Assessing the Real-world Performance of Hearing Aids using Mobile Phones

2:15pm

Sayan Bandyapadhyay

Digging Deeper into k-means Clustering

2:35-2:45pm

Break

2:45-3:45pm

Session Two

2:45pm

Zhe Li

Open the Magic Box of Deep Learning for Image Classification

3:05pm

Talal Riaz

Distributed Maximal Independent Set Problem

3:25pm

Pooya Rahimian

Using a Virtual Environment to Study the Impact of Sending Traffic Alerts to Texting Pedestrians.

3:45-4:00pm Break
4:00-5:00pm

Keynote Speaker Alberto Segre

Current Research in Computational Epidemiology

5:00-5:30pm

Reception

View on UI Event Calendar

Speakers

Baoluo - UI CS PhD candidateBaoluo Meng

Title: Relational Constraints Solving via SMT

Abstract: Many computational problems require reasoning about relational structures such as system high-level design, architectural configurations of network systems, ontology, and verification of programs with linked data structure. Relational logic is an appealing choice for reasoning about such problems. Alloy is a declarative modeling language based on first-order relational logic with built-in transitive closure. It is designed to model structurally-rich problems. The analysis of Alloy models is performed automatically by the Alloy Analyzer - a SAT-based finite model finder. However, Alloy is limited in two major aspects: i) it can only disprove properties of input models, not prove them ii) it has limited support for numerical constraints. SMT solvers are efficient automated tools that can check the satisfiability of complex constraints over several domains, including arithmetic and the theory of finite sets. In this talk, I will present a novel approach for addressing the limitations of the Alloy Analyzer by leveraging the power of SMT solvers.

4th year PhD student | Advisor: Cesare Tinelli | Area of research: Formal Methods


Shabih Hasan - UI CS PhD CandidateShabih Hasan

Title: Assessing the Real-world Performance of Hearing Aids using Mobile Phones

Abstract: Hearing aids (HAs) are an essential tool for helping individuals with hearing loss lead a normal life. A significant portion of individuals that may benefit form using HAs do not use them, and the satisfaction of those who do is only between 60-65%. The traditional methodologies for evaluating HA performance provide a limited understanding regarding the factors associated with the low success rate because they cannot predict the real-world performance of HAs. The core of the problem lies in the fact that laboratory based measurements cannot fully reproduce real-world environments within the confines of laboratory spaces. In this talk I shall present my work, AudioSense, which is a mobile computing solution for evaluating hearing aids in-situ and in real-time. AudioSense has provided us with the ability to characterize the real-world acoustic contexts that a HA user experiences in their day to day lives. In addition to this I shall present the models we have built to predict HA success using the data we have collected with AudioSense.

5th year PhD student | Advisor: Octav Chipara | Area of research: Mobile Computing


Sayan Bandyapadhyay - UI CS PhD CandidateSayan Bandyapadhyay

Title: Digging Deeper into k-means Clustering

Abstract: Clustering problems often arise in fields like data mining, machine learning, and so on. Clustering usually refers to the task of partitioning a collection of objects into groups with similar elements, with respect to a similarity (or dissimilarity) measure. Among the clustering problems, k-means clustering in particular has received much attention from researchers. Despite the fact that k-means is a well studied problem, its status in the plane is open. In particular, it is unknown whether one can get a "near optimal" solution for this problem in the plane in polynomial time. 
In this talk, I am planning to describe a simple local search based algorithm for this problem. The algorithm may use a little bit more than k clusters, but produces a "near optimal" solution in polynomial time. This is joint work with Kasturi Varadarajan, and has appeared recently in the 32nd International Symposium on Computational Geometry (SoCG 2016).

4th year PhD student | Advisor: Kasturi R. Varadarajan | Area of research: Computational Geometry


Zhe Li - UI CS PhD CandidateZhe Li

Title: Open the Magic Box of Deep Learning for Image Classification

Abstract: Deep neural network achieves the tremendous success in a variety of application areas such as image classification, speech recognition, natural language processing among many others. Besides its successful story, deep neural network itself is not well understood by a broad audience and is considered as the magic box. In this talk, I will open this magic box of deep neural network, present what is inside, and discuss why deep neural network works well through the successful example especially for image classification. In the end, I will briefly discuss our proposed technique that not only speeds up the training of deep neural network but also improves the generalization performance of the model.

4th year PhD student | Advisor: Tianbao Yang | Area of research: Machine Learning


Talal Riaz - UI CS PhD candidateTalal Riaz

Title: Distributed Maximal Independent Set Problem

Abstract: In the field of distributed algorithms, there has been considerable interest in the finding fastest possible algorithms for solving classical graph problems. In this regards, the Maximal Independent Set (MIS) problem has been a prime focus for researchers and has even been referred to as the "crown jewel" of distributed computing. Researchers in the 80s independently came up with a simple randomized algorithm (commonly referred to as Luby’s algorithm) to solve the problem in O(log n) communication rounds (where n refers to the number of nodes in the graph). This algorithm, to this day, remains the state of the art in terms of complexity with respect to n. Nevertheless, researchers have recently improved upon Luby’s algorithm in restricted graph classes such as trees, and in cases where the maximum degree in the graph is some slowly growing function of n. In this talk, we will discuss the maximal independent set problem, as well as our improvement beyond Luby’s algorithm in bounded arboricity graphs.

4th year PhD student | Advisor: Sriram Pemmaraju | Area of research: Distributed Algorithms


Pooya Rahimian - UI CS PhD CandidatePooya Rahimian

Title: Using a Virtual Environment to Study the Impact of Sending Traffic Alerts to Texting Pedestrians.

Abstract: This paper presents an experiment conducted in a large-screen virtual environment to evaluate how texting pedestrians respond to permissive traffic alerts delivered via their cell phone. We compared gap selection and movement timing in three conditions: texting, texting with alerts, and no texting (control). Participants in the control and alert groups chose larger gaps, were more discriminating in their gap choices, and had more time to spare than participants in the texting group. The alert group also paid the least attention to the roadway. The results demonstrate the potential and risks of Vehicle-to-Pedestrian (V2P) communications technology for mitigating pedestrian-vehicle crashes.

4th year PhD student | Advisor: Joseph K. Kearney | Area of research: Virtual Environments


Alberto Segre - UI CS Professor, Chair and Gerard P Weeg Faculty Scholar in Informatics

 

Keynote Speaker Alberto Segre

Title: Current Research in Computational Epidemiology

Abstract: In this talk, I will describe current work in the University of Iowa Computational Epidemiology group (compepi.cs.uiowa.edu), where our research involves the use of computational tools and technology to quantify, simulate, visualize and, in general, better understand the spread of disease. Our goal is to inform public and hospital policy decisions with respect to disease surveillance, disease prevention measures, and outbreak containment.

Bio: Alberto Maria Segre is Professor and Chair of the Computer Science Department and Gerald P. Weeg Faculty Scholar in Informatics at the University of Iowa. He holds a B.A. in Music Theory and a B.S., M.S. and Ph.D. in Electrical Engineering, all from the University of Illinois at Urbana-Champaign. His research interests focus on distributed algorithms for discrete optimization problems, with emphasis on algorithmic problems in the biological and health sciences. Most recently, his work has focused on epidemiological simulation and modeling.


 

Directions

 MacLean Hall and Pappajohn Business Building

 

 

Getting to the Conference

Iowa City lies just off of Interstate 80 in Eastern Iowa. Regardless of whether you are coming from the West or the East, you will want to take exit 244 off of I-80 following Dubuque Street until you intersect with Church Street. Follow Church Street as it turns into Clinton Street and continue straight until you reach the downtown area and the Pentacrest (you will see the Old Capitol to the West), which is the heart of the University of Iowa's campus. All of the symposium will take place in that general vicinity in the Pappajohn Business Building and MacLean Hall. For a visual representation, see the map.