<?xml version="1.0" encoding="ISO-8859-1" ?>
<rss version="2.0"
     xmlns:rdf="http://www.w3.org/1999/02/22-rdf-syntax-ns#"  
     xmlns:ev="http://purl.org/rss/1.0/modules/event/" 
     xmlns:dc="http://purl.org/dc/elements/1.1/"
>
 <channel>
  <title>URCS Seminars &amp; Talks</title>
  <link>http://www.cs.rochester.edu/dept/seminar</link>
  <description>Upcoming Seminars &amp; Talks from the Computer Science Department at the University of Rochester</description>
  <language>en-us</language>
  <image>
   <url></url>
   <title>Computer Science Department, University of Rochester</title>
   <link>http://www.cs.rochester.edu/dept/seminar</link>
  </image>
  <lastBuildDate>Thu, 09 Feb 2012 11:16:08 -0500</lastBuildDate>
  
   <item>
    <title>Naushad UzZaman: Interpreting the Temporal Aspects of Language </title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/728</link>
    <description>[Friday, February 10, 2012 at 12:45 PM in CSB 703] Abstract

Understanding temporal information in the text is fundamental for deep language understanding and key to many advanced NLP applications, such as question answering, information extraction, timeline visualization, and document summarization. These techniques can be applied in news, medical, history and other domains.

In this talk, I will present our hybrid system to automatically extract temporal information from raw text by extracting events, temporal expressions and identifying temporal relations between them. Our system had a competitive performance in the temporal evaluation shared task - TempEval 2010. Next, I will present a metric that we developed for evaluation of temporal annotation. Our metric has been adopted by the premier temporal evaluation shared task, TempEval 2013, for evaluating participating systems. Finally, I will present a question-answering system that can answer temporal questions with temporal reasoning. Our developed QA system can also be used to evaluate temporal information understanding capability.

I will also briefly talk about my other projects, ranging from multimodal summarization of complex sentence to game prediction using social media.

Bio: Naushad UzZaman is a PhD candidate under Professor James F. Allen in the Computer Science department at the University of Rochester (URCS). His research interests are in Natural Language Understanding (NLU) and Natural Language Processing (NLP), with focus on Temporal Information Processing, Information Extraction, Social Media Text Analysis, Medical NLP, Question Answering, and Multimodal Summarization. At URCS, he primarily worked on temporal information processing. He is co-organizing the TempEval 2013 shared task. He also worked on making information accessible with Jeffrey Bigham. He has done multiple research internships in various domains, such as, game prediction using social media (Yahoo! Research), medical NLP (Microsoft Medical Media Lab), and car dialog systems (Bosch Research and Technology Center).</description>
    <ev:startdate>20120210T174500Z</ev:startdate>
    <ev:enddate>20120210T174500Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/728</guid>
   </item>
  
   <item>
    <title>Jonathan Gordon, Lenhart Schubert: Gogate &amp; Domingos: Probabilistic Theorem Proving.</title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/729</link>
    <description>[Friday, February 10, 2012 at  3:30 PM in CSB 703] http://www.hlt.utdallas.edu/~vgogate/papers/uai11-b.pdf</description>
    <ev:startdate>20120210T203000Z</ev:startdate>
    <ev:enddate>20120210T220000Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/729</guid>
   </item>
  
   <item>
    <title>Tom Yeh: Making GUI Automation Accessible using Computer Vision</title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/720</link>
    <description>[Friday, February 17, 2012 at 11:00 AM in 209 Computer Studies Building] Most people dislike manual and repetitive tasks and like to automate them if the right tool is available. For GUI automation, the right tool does not always exist. Some require programming and are inaccessible for end-users who have little knowledge about programming. Some require interacting with a GUI’s internal structure and are unable to deal with proprietary and legacy applications whose internal structure is inaccessible. To make automation accessible, we need to find a new modality that is commonly available in all GUI applications and easily understood by end-users. One such modality I have tried with great success is vision. In this talk, I will introduce computational techniques that use images of GUI applications as first-class objects to allow end-users to automate any GUI application they see on a computer screen. I will present Sikuli, software I created that has enabled tens of thousands of users to automate repetitive tasks they were unable to automate before. I will show many real uses of Sikuli such as automating daily disk cleanup, automating a complex sign-up process, automating Facebook status updates, automating dialing on an Android phone, and automating Angry Birds. I will illustrate the real benefit of automation with case studies such as the one about a software project that uses Sikuli to automate 400+ previously manual tests, doubling the software’s release rate. I will discuss lessons learned from Sikuli&#39;s user community and new research problems it has inspired. Finally, I will outline key challenges for future research to make automation accessible for the entire lifecycle of software including design, development, testing, use, and support.

Bio: Tom Yeh is an assistant research scientist in the University of Maryland Institute for Advanced Computer Studies (UMIACS). He received his Ph.D. degree in Computer Science at MIT in 2009. He then spent two years doing a postdoc at the University of Maryland College Park. His research interests span human computer interaction, computer vision, and software engineering. He has written over 30 research publications on algorithms for interactive computer vision, vision-based interactive systems, multimedia information retrieval, and visual software test automation. He has served on the program committees of the conferences in his area including the Symposium on User Interface Software and Technology and the Workshop on Compute Vision Application. He has been awarded Best Student Paper at UIST 2009, Best Paper at UIST 2010. He earned his master’s degree in Computer Science at MIT, and a bachelor’s degree in Computer Science at Simon Fraser University.


</description>
    <ev:startdate>20120217T160000Z</ev:startdate>
    <ev:enddate>20120217T170000Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/720</guid>
   </item>
  
   <item>
    <title>Anne Marie Piper: Cooperative Multimodal Interaction:  Designing for Communication, Learning, and Social Behavior</title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/722</link>
    <description>[Monday, February 20, 2012 at 11:00 AM in 209 Computer Studies Building] Increasingly, our interactions with other people are mediated by technology. Even when collocated with another person, computer systems shape how we share information, coordinate our activities, and communicate. Systems that enable gestural interaction (i.e., through multitouch sensing or pen-based techniques) and speech input not only increase the expressiveness of computer interfaces, but also open up new avenues for augmenting how people work together and share digital information. My research in human-computer interaction focuses on multimodal computer interfaces to support collocated human activity. In this talk I will present several gesture- and speech-based interfaces I have designed with collaborators to address challenges in communication, learning, and social interaction. I apply theories of cognitive science to inform the design of multimodal systems, and in turn, develop new insights about human cognitive and social behavior through observations of system use. I examine how cooperative multimodal interaction promotes reflection on spoken and written information and increases one&#39;s ability to communicate with others. I will demonstrate how multimodal interfaces can bridge interaction challenges for individuals with a hearing, speech, or social disability as well as improve information access for people of all ages and abilities.

Bio: Anne Marie Piper is a lecturer and researcher in Cognitive Science at the University of California, San Diego. She teaches courses in human-computer interaction and interaction design. Anne Marie received her PhD in Cognitive Science from UCSD in 2011. Over the past ten years she has researched how new computer interfaces support learning activities and the communication needs of individuals with disabilities. Anne Marie received her bachelor’s degree in Computer Science from the Georgia Institute of Technology and her masters in Education from Stanford University. She is the recipient of an NSF Graduate Research Fellowship, the 2010 Best Student Paper Award at the ACM Conference on Computers and Accessibility (ASSETS), and a 2011 UCSD Interdisciplinary Scholar Award. Anne Marie has also worked as a user experience researcher at Microsoft and LeapFrog.</description>
    <ev:startdate>20120220T160000Z</ev:startdate>
    <ev:enddate>20120220T170000Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/722</guid>
   </item>
  
   <item>
    <title>Yizhou Sun: Mining Heterogeneous Information Networks</title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/723</link>
    <description>[Monday, March 05, 2012 at 11:00 AM in 209 Computer Studies Building] The analysis of social and information networks has gained wide attentions in computer science, physics, social science, biology, and other research communities, with exciting discoveries and successful applications. In most of the existing network research, networks are usually assumed to be homogenous, where nodes are objects from the same entity type (e.g., person) and links are relationships from the same relation type (e.g., friendship). However, in reality, objects of different types interact with each other via relationships of different types, forming heterogeneous, semi-structured information networks. Such kind of heterogeneous networks are ubiquitous, representing real-world systems ranging from social to scientific, engineering, or medical systems, and to e-commerce systems.
 
We investigate the problem of mining heterogeneous information networks, by leveraging the semantic meanings of the types of objects and links in the network, and propose principles and methodologies that can exploit these rich semantics and solve large-scale, real-world problems. In particular, we disclose how different types of relationships carry different semantics and strengths in determining the similarity or influence between linked objects. These studies have laid the foundation for in-depth analysis of heterogeneous information networks, such as similarity search, ranking, clustering, classification, prediction and outlier detection. Our experiments on large-scale networks like the DBLP bibliographic network, the Flickr image network, and the Yelp review network, have demonstrated the effectiveness of our models and the efficiency of our algorithms, as well as the potential of our methodologies being successfully applied to a broader range of applications.

Bio: Yizhou Sun is a fifth year Ph.D. candidate at the Department of Computer Science, University of Illinois at Urbana-Champaign. Her principal research interest is in mining information and social networks, and more generally in data mining, database systems, statistics, machine learning, information retrieval, and network science, with a focus on modeling novel problems and proposing scalable algorithms for large-scale, real-world applications. Yizhou has over 30 publications in book chapters, journals, and major conferences such as SIGKDD, VLDB, NIPS and so on. Tutorials based on her thesis work on mining heterogeneous information networks have been given in several premier conferences. A contracted monograph “Mining Heterogeneous Information Networks: Principles and Methodologies” is to be published in 2012.
</description>
    <ev:startdate>20120305T160000Z</ev:startdate>
    <ev:enddate>20120305T170000Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/723</guid>
   </item>
  
   <item>
    <title>Professor Adrien Treuille: G. Milton Wing Lectures : Interactive Biology</title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/701</link>
    <description>[Wednesday, March 07, 2012 at  5:00 PM in Hutchison Hall  Room 140(Hubbell Auditorium)] Nanoscale engineering holds the key to developing next-generation therapies, but this painstaking process depends on the difficult task of predicting molecular self-assembly. This talk describes two unprecedented &quot;citizen science&quot; projects to rapidly advance bioengineering. Our on-line protein folding game, Foldit, challenges non-experts to fold proteins, a puzzle akin to &quot;3D Tetris.&quot; Our brand new nano-engineering game, EteRNA, pushes this concept into the rich and confounding world of real experimentation and validation: every week, we synthesize top EteRNA player designs and return this data as score back into the game. EteRNA thus enabling thousands of enthusiasts to stringently test scientific hypotheses on a weekly basis. Both games now hosts a thriving communities of citizen scientists who are outperforming existing state-of-the-art algorithmic methods.

Bio: Dr. Treuille, who received his Ph.D. from the University of Washington in 2008,
is one of the country&#39;s leading young computer scientists. Currently on the faculty of Carnegie Mellon University,
he  holds an NSF CAREER Award, is a panelist for the President&#39;s Council of Advisors on Science and Technology, and in 2009 was named by MIT Technology Review as one of the Top  Innovators Under 35. </description>
    <ev:startdate>20120307T220000Z</ev:startdate>
    <ev:enddate>20120307T230000Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/701</guid>
   </item>
  
   <item>
    <title>Professor Adrien Treuille: G. Milton Wing Lectures: New Approaches to Modeling and Control of Complex Dynamics</title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/702</link>
    <description>[Thursday, March 08, 2012 at  4:30 PM in Goergen Hall 101 (Sloan Auditorium)] Complex phenomena such as animal morphology, human motion, and large fluid systems challenge even our most sophisticated simulation and control techniques. My overarching research goal has been to develop fundamentally new methods to approach such high-dimensional and nonlinear problems. This talk presents my work solving these problems across a wide range of phenomena, including a new model-reduction approach to fluids that is orders-of-magnitude faster than standard simulation methods and enables interactive high-resolution fluid simulation for the first time. Another example is a continuum approach to crowd dynamics which efficiently reproduces empirical aspects of large crowd behavior that would be difficult or impossible to achieve with traditional agent models. The talk will also cover work on several other phenomena including human animation, animal morphology, and protein folding. Such new algorithmic approaches advance not only our ability to simulate and control complex systems but also our understanding of the systems themselves.

 Bio: Dr. Treuille, who received his Ph.D. from the University of Washington in 2008,
is one of the country&#39;s leading young computer scientists. Currently on the faculty of Carnegie Mellon University,
he  holds an NSF CAREER Award, is a panelist for the President&#39;s Council of Advisors on Science and Technology, and in 2009 was named by MIT Technology Review as one of the Top  Innovators Under 35. 
</description>
    <ev:startdate>20120308T213000Z</ev:startdate>
    <ev:enddate>20120308T223000Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/702</guid>
   </item>
  
   <item>
    <title>Patrick Gage Kelley: TBA</title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/724</link>
    <description>[Monday, March 19, 2012 at 11:00 AM in 209 Computer Studies Building] Abstract</description>
    <ev:startdate>20120319T150000Z</ev:startdate>
    <ev:enddate>20120319T160000Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/724</guid>
   </item>
  
   <item>
    <title>Duen Horng Chau: Data Mining Meets HCI: Making Sense of Large Graphs</title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/725</link>
    <description>[Monday, March 26, 2012 at 11:00 AM in 209 Computer Studies Building] We have entered the era of big data. Datasets surpassing terabytes now arise in science, government and enterprises. Yet, making sense of these data remains a fundamental challenge. Where to we start our analysis? Where to go next? And how to visualize our findings? My research takes a step towards answering these questions.

I work in Data Mining and Human-Computer Interaction (HCI), and I combine the best from both worlds to create tools that help people make sense of graphs with billions of nodes and edges. I present my work in three interrelated topics:

(1) Attention Routing: I introduce this idea, based on anomaly detection and machine inference, that automatically draws people&#39;s attention to interesting parts of the graph. I describe two examples: the Polonium technology unearths malware from 37 billion machine-file relationships; the NetProbe system fingers bad guys who commit auction fraud.

(2) Mixed-Initiative Graph Sensemaking: I describe the Apolo system that combines machine inference and visualization to guide the user to interactively explore large graphs. The user gives examples of relevant nodes, and Apolo recommends which areas the user may want to see next.

(3) Scaling Up: I show how we may enable interactive analytics of large graphs with a hybrid architecture that harnesses parallel computation for expensive tasks, and local computation for fast machine inference, visualization, and interaction.

Bio: Duen Horng &quot;Polo&quot; Chau is a Ph.D. candidate in the Machine Learning Department at Carnegie Mellon University. He received a Masters in Human-Computer Interaction (HCI) from Carnegie Mellon.

Polo is working to bridge the fields of Data Mining and HCI. He develops tools that combine the best of both worlds to help people make sense of large graphs with billions of nodes and edges. His research interests span data mining, machine learning, information visualization and HCI.

Polo solves large-scale problems that bring great impact to society. His NetProbe auction fraud detection system made headlines on The Wall Street Journal, CNN, TV and radio. His Polonium malware detection technology (with Symantec, patent-pending) protects 120 million people worldwide.

Polo is the only two-time Symantec fellow. He received a Yahoo! Key Scientific Challenges Award. He contributes to the PEGASUS peta-scale graph mining that won an Open Source Software World Challenge Silver Award. Polo is also an award-winning designer. He designed Carnegie Mellon&#39;s ID card.</description>
    <ev:startdate>20120326T150000Z</ev:startdate>
    <ev:enddate>20120326T160000Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/725</guid>
   </item>
  
   <item>
    <title>Alexandra Meliou: Why and How: A Reverse Perspective on Data Management</title>
    <link>http://www.cs.rochester.edu/dept/seminar/view/726</link>
    <description>[Monday, April 02, 2012 at 11:00 AM in 209 Computer Studies Building] Current trends have seen data grow larger, more intertwined, and more diverse, as more and more users contribute to and use it. This trend has given rise to the need to support richer data analysis tasks. Such tasks involve determining the causes of observations, finding and correcting the sources of error in query results, as well as modifying the data in order to make it conform to complex desirable properties.

In this talk I will discuss three challenges: (a) providing explanations through support for causal queries (&quot;Why&quot;), (b) tracing and correcting errors at their source (post-factum data cleaning), and (c) integrating database systems with constrained optimization capabilities (&quot;How&quot;). First, I will show how to apply causal reasoning to tuple provenance in order to determine the causes of query results, and their responsibility. I will present extensive analysis of the data complexity for the case of conjunctive queries, and focus on a complete dichotomy between NP-hard and PTIME cases for the problem of computing responsibility. This concrete characterization of PTIME cases is crucial in scaling up to the challenges of Big Data. Second, I will demonstrate the applicability of the causality framework in a practical setting. I will use a mobile sensing application to show that ranking provenance tuples by their degrees of responsibility identifies errors more effectively than other schemes. Finally, I will present the Tiresias system, the first how-to query engine, which seamlessly integrates database systems with constrained problem solving capabilities. The contributions of the system are threefold: (a) a declarative interface for defining how-to queries over a database, (b) translation rules from the declarative statements to the constrained problem specification, and (c) a suite of data-specific optimizations that allow scaling to large data sizes. Initial results of our prototype system implementation show order-of-magnitude speedups to state-of-the-art solver runtimes, which indicates that there are significant gains in pushing this functionality within the database engine. 
I will conclude with a summary of my contributions, and discuss my future steps with the Tiresias system, and the bigger vision of reverse data management.

Bio: Alexandra Meliou is a postdoctoral research associate with Dan Suciu in the database group of the University of Washington. She received her Ph.D degree in 2009 from the University of California, Berkeley, and is a 2008 Siebel Scholar. Her interests are in data and information management with a focus on issues of data provenance. Currently, she is working on extending the capabilities of database systems to support business decisions and strategy planning queries.
</description>
    <ev:startdate>20120402T150000Z</ev:startdate>
    <ev:enddate>20120402T160000Z</ev:enddate>
    <guid>http://www.cs.rochester.edu/dept/seminar/view/726</guid>
   </item>
  
 </channel>
</rss>

