Select a calendar:
Filter April Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Conferences, Lectures, & Seminars
Events for April
-
Harshvardhan Vathsangam, USC (PhD Defense)
Mon, Apr 01, 2013 @ 02:30 PM - 04:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Harshvardhan Vathsangam, USC
Talk Title: Sense and Sensibility: Statistical Techniques for Human Energy Expenditure Estimation Using Kinematic Sensors
Series: PhD Defense Announcements
Abstract: Sense and Sensibility: Statistical Techniques for Human Energy Expenditure Estimation Using Kinematic Sensors ==== Healthcare is undergoing a shift from the episodic, expert-driven, curative approaches of the past towards a self-empowered, preventative model for the future. Central to this is the treatment of chronic illnesses. This treatment will require the adoption of behavioral changes on one's lifestyle. In this thesis, we focus on the negative effects of one such chronic illness: physically inactivity.
Regular physical activity is associated with decreased mortality, lower risk of cardiovascular disease, diabetes mellitus, colon and breast cancer.
Despite this knowledge, physical activity levels are not adequate.
Central to the need to get people to be more active is the ability to accurately measure and characterize physical activity in a cost-effective yet ubiquitous manner. One dimension of characterization of physical activity is the energy expended as a result of that activity. In this dissertation, we aim to demonstrate how kinematic sensors in combination with statistical techniques can accurately predict energy expenditure due to physical activity.
We cast the problem of determining energy expenditure in a mathematical framework and discuss various functional maps. We derive a set of frequency-based features that are robust to location on the human body and orientation. We use these features to determine the most accurate 'per-person' technique to map movement to energy expenditure. We compare prediction accuracies using different sensor streams and algorithms. A comparative study of accuracy versus inference time is also performed. We extend this work to be able to generate maps given a minimal set of morphological descriptors such as height, weight, age etc. We present and compare a set of models including nearest neighbor models, weight-scaled models, a set of hierarchical linear models and speed-based approaches. We show how these approaches can be used to evaluate the best subset of morphological descriptors and the best individual descriptor to generate personalized maps across people. These contributions are a step towards designing cost-effective, accurate and ubiquitous solutions to estimate physical activity levels and designing interventions based on accurately measured data.
Host: Lizsl DeLeon
Location: Ronald Tutor Hall of Engineering (RTH) - 406
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: Mehmet Dogar (CMU): Physics-Based Robotic Manipulation in Human Environments
Tue, Apr 02, 2013 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Mehmet Dogar, Carnegie Mellon
Talk Title: Physics-Based Robotic Manipulation in Human Environments
Series: CS Colloquium
Abstract: The list of physics-based actions that we humans use to push, pull, throw, tumble, and play with the objects around us is nearly endless.
My research strives to develop robots with similar capabilities by incorporating physical predictions into manipulation planning. Most existing manipulation planners do not use physical predictions and therefore are limited to pick-and-place actions. I develop manipulation algorithms which enable robots to move beyond pick-and-place.
In this talk I will focus on using physics-based pushing actions. I will describe how a robot can plan pushing actions that are robust to high degrees of uncertainty in the environment. I will show that pushing manipulation leads to very efficient plans in cluttered environments, while pick-and-place manipulation treats clutter like a game of chess where each piece is moved one-by-one. Finally, I will talk about how contact sensor feedback can be used during physics-based actions to reduce uncertainty and to account for errors in the robot's physical predictions.
Biography: Mehmet Dogar is a PhD student at the Robotics Institute at Carnegie Mellon University. His research focuses on using physics-based predictions in robotic manipulation which enables robots to accomplish useful tasks in dynamic and cluttered human environments. He received his M.S. and B.S. degrees in Computer Engineering from the Middle East Technical University, Turkey. He was a Finalist for the Best Paper Award at IROS 2010. He received the Fulbright Award for Outstanding Foreign Students in 2008.
Host: Stefan Schaal
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: Stefano Tessaro (MIT CSAIL): Theoretical Foundations for Applied Cryptography
Wed, Apr 03, 2013 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Stefano Tessaro, MIT CSAIL
Talk Title: Theoretical Foundations for Applied Cryptography
Series: CS Colloquium
Abstract: My talk explains that obtaining quality applied cryptography (i.e., with the desired combination of security assurance and performance) requires significant and deep advances in theory. I will discuss three illustrative examples.
First, I will present my results on a process called security amplification that may be used to make block ciphers (the workhorses of modern cryptography under which encryption is ubiquitously
performed) more secure against cryptanalysis.
Second, I will introduce my theory of multi-instance security, which may be applied to provide the first theoretical analysis of the effectiveness of the classical practice of password salting.
Third, I will bridge a 35-year gap between the information & coding community and the cryptography community by providing cryptographic foundations, as well as schemes with optimal parameters, for private communication based solely on the assumption that the communication channel from sender to adversary is noisier than the one from sender to receiver. The resulting schemes, being keyless, are particularly attractive in wireless communication scenarios.
Biography: Stefano Tessaro is currently a research scientist in the Cryptography and Information Security group at MIT CSAIL. He received his MSc and PhD from ETH Zurich in 2005 and 2010, respectively. From 2010 to 2012, he was a postdoctoral scholar at the University of California, San Diego. His research interests are in cryptography and its connections to theoretical computer science and information theory.
Host: David Kempe
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium Series Lecture: Greg ver Steeg (ISI): Detecting Influence in Social Networks
Thu, Apr 04, 2013 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Greg ver Steeg, Information Sciences Institute (ISI)
Talk Title: Detecting Influence in Social Networks
Series: CS Colloquium
Abstract: Current tests for contagion in social network studies are flawed due to the confounding effects of latent homophily (i.e., ties form preferentially between individuals with similar hidden traits). We demonstrate a general method to lower bound the strength of causal effects in observational social network studies, even in the presence of arbitrary, unobserved individual traits. Our test requires no parametric assumptions and is based on connections with algebraic geometry and Bell inequalities in quantum physics. We demonstrate the effectiveness of our approach by correctly deducing the causal effects for examples previously shown to expose defects in existing methodology. Finally, we discuss preliminary results on data taken from the Framingham Heart Study showing that the spread of obesity cannot be explained by latent homophily.
I will also give a brief summary of other recent and ongoing research. One goal is to construct an information-theoretic foundation for unsupervised learning with preliminary success in discovering predictable relationships in social networks based only on content dynamics. Other topics include the statistical physics of networks and solving machine learning problems with quantum annealing.
Biography: Greg Ver Steeg is a computer scientist at USC's Information Sciences Institute. Ver Steeg received his PhD in physics from Caltech in 2009 for research in quantum information theory. His research has explored a diverse set of connections between computer science and physics dealing with machine learning, information theory, causal inference, and information processing. He is the recent winner of an AFOSR Young Investigator Award.
Host: CS Colloquium Series Lecture
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
Dian Gong (USC), Student Seminar Series
Mon, Apr 08, 2013 @ 12:00 PM - 02:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Dian Gong, USC Electrical Engineering Ph.D. Student
Talk Title: Machine Learning to Structured Time Series Analysis
Series: Student Seminar Series
Abstract: Time series and sequential data have been investigated for several decades in statistics, signal processing and econometrics. The success of machine learning techniques brings opportunities to efficiently analyze more complex, high dimensional and large-scale time series data. In this talk, we present a non-parametric learning framework to multivariate time series data. The raw time series is first temporally decomposed into different units with certain semantic meanings by our newly proposed Kernelized Temporal Cut (KTC). KTC is an online non-parametric change-point detection method that can detect regime changes for complex sequential data. Given two time series units, the similarity (distance) can be calculated by using spaito-temporal alignment methods such as DTW. To handle the nonlinearity of multivariate time series data in many applications, we propose Dynamic Manifold Warping (DMW). DMW is a combination of DTW and manifold learning by exploring the intrinsic manifold structure of time series data. After temporal segmentation and the design of distance metric, we can treat each time series unit as one data instance, and many tasks can be performed, such as clustering, classification and retrieval. We apply this framework to human action analysis and achieve promising results.
Biography: Dian Gong is a PhD Candidate major in Electrical Engineering and minor in Computer Science at USC. His advisor is Prof. Gerard Medioni, and his research areas are machine learning to structured time series, manifold learning and probabilistic Tensor Voting. He also uses cutting-edge machine learning techniques to computer vision applications such as human activity recognition. His works are published at conferences such as AISTATS 2010, ICCV 2011, ICML 2012 and ECCV 2012. He worked as a summer quantitative trading associate at Barclays Capital New York office. He also worked as research intern at Sony US Research, San Jose, CA on distance metric learning, and Microsoft Research Asia, on probabilistic graphical model. In the past, he won several mathematics contest awards such as the Gold medal of China Mathematics Olympiad, and selected as national team candidate for International Mathematics Olympiad. He got his BS in Electronic Engineering from Tsinghua University.
Advisor: Gerard Medioni
Refreshments will be provided
Host:
More Info: https://mhi.usc.edu/events/event-details/?event_id=901927
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
Event Link: https://mhi.usc.edu/events/event-details/?event_id=901927
-
CS Colloquium: Lorenzo Torresani (Dartmouth): Challenges and Opportunities in Visual Recognition with Big Image Data
Mon, Apr 08, 2013 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Lorenzo Torresani, Dartmouth
Talk Title: Challenges and Opportunities in Visual Recognition with Big Image Data
Series: CS Colloquium
Abstract: The last few years have seen a tremendous explosion of image and video data on the Web. Unfortunately only a small portion of this visual data is annotated with text. Even when tags are available, they often do not describe accurately the semantics of the image or the video. This renders traditional text-search an ineffective tool on these collections. In this talk I will describe some of my recent work on designing visual recognition systems that can help users browse and search image repositories more effectively.
I will begin with an algorithm that addresses the computational challenges posed by visual recognition in Web-scale image databases. Our approach centers around the learning of a compact image code optimized to yield accurate recognition with linear (i.e., efficient) classifiers: even when the representation is compressed to less than 300 bytes per image, linear classifiers trained on our descriptor yield accuracy matching the state-of-the-art but at orders of magnitude lower computational cost.
In the second part of my talk I will present a method that embraces Big Image Data as an opportunity to improve visual recognition. Our algorithm exploits a dataset of 10 million labeled photos to learn a universal semantic distance between images. This metric can be used either as a similarity measure to find pictures by example or as a ââ¬Åkernelââ¬Â in distance-based image classifiers, yielding a significant boost in accuracy over traditional metrics.
Biography: Lorenzo Torresani is an Assistant Professor in the Computer Science Department at Dartmouth College. He received a Laurea Degree in Computer Science with summa cum laude honors from the University of Milan (Italy) in 1996, and an M.S. and a Ph.D. in Computer Science from Stanford University in 2001 and 2005, respectively. In the past, he has worked at several industrial research labs including Microsoft Research Cambridge, Like.com, and Digital Persona. His research interests are in computer vision and machine learning. In 2001, Torresani and his coauthors received the Best Student Paper Award at the IEEE Conference On Computer Vision and Pattern Recognition (CVPR). He is the recipient of a National Science Foundation CAREER Award.=
Host: Gerard Medioni
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: Aditya Parameswaran (Stanford): Human-Powered Data Management
Tue, Apr 09, 2013 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Aditya Parameswaran, Stanford
Talk Title: Human-Powered Data Management
Series: CS Colloquium
Abstract: Fully automated algorithms are inadequate for a number of data analysis tasks, especially those involving images, video, or text. Thus, there is a need to combine "human computation" (or crowdsourcing), together with traditional computation, in order to improve the process of understanding and analyzing data. My thesis addresses several topics in the general area of human-powered data management. I design algorithms and systems for combining human and traditional computation for: (a) data processing, e.g., using humans to help sort, cluster, or clean data; (b) data extraction, e.g., having humans help create structured data from information in unstructured web pages; and (c) data gathering, i.e., asking humans to provide data that they know about or can locate, but that would be difficult to gather automatically. My focus in all of these areas is to find solutions that expend as few resources as possible (e.g., time waiting, human effort, or money spent), while still providing high quality results.
In this talk, I will first present a broad perspective of our research on human-powered data management, and I will describe some systems and applications that have motivated our research. I will then present details of one of the problems we have addressed: filtering large data sets with the aid of humans. Finally I will argue that human-powered data management is an area in its infancy, by describing a number of open problems I intend to address in my future research program.
Biography: Aditya Parameswaran is a Ph.D. student in the InfoLab at Stanford University, advised by Prof. Hector Garcia-Molina. He is broadly interested in data management, with research results in human computation, information extraction, and recommendation systems. Aditya is a recipient of the Key Scientific Challenges Award from Yahoo! Research (2010), two best-of-conference citations (VLDB 2010 and KDD 2012), and the Terry Groswith graduate fellowship at Stanford University.
Host: Cyrus Shahabi
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
Manish Jain, USC (PhD Defense)
Wed, Apr 10, 2013 @ 01:00 PM - 03:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Manish Jain, USC
Talk Title: Thwarting Adversaries with Unpredictability: Massive-scale Game-Theoretic Algorithms for Real-world Security Deployments
Series: PhD Defense Announcements
Abstract: Protecting critical infrastructure and targets such as airports, transportation networks, power generation facilities as well as critical natural resources and endangered species
is an important task for police and security agencies worldwide. Securing such potential targets using limited resources against intelligent adversaries in the presence of the uncertainty and complexities of the real-world is a major challenge. My research uses a game-theoretic framework to model the strategic interaction between a defender (or security forces) and an attacker (or terrorist adversary) in security domains.
Game theory provides a sound mathematical approach for deploying limited security resources to maximize their effectiveness. While game theory has always been popular in the arena of security, unfortunately, the state of the art algorithms either fail to scale or to provide a correct solution for large problems with arbitrary scheduling constraints. For example, US carriers fly over 27,000 domestic and 2,000 international flights daily, presenting a massive scheduling challenge for Federal Air Marshal Service (FAMS).
My thesis contributes to a very new area that solves game-theoretic problems using insights from large-scale optimization literature towards addressing the computational challenge posed by real-world domains. I have developed new models and algorithms that compute optimal strategies for scheduling defender resources is large real-world domains. My thesis makes the following contributions. First, it presents new algorithms that can solve for trillions of actions for both the defender and the attacker. Second, it presents a hierarchical framework that provides orders of magnitude scale-up in attacker types for Bayesian Stackelberg games. Third, it provides an analysis and detection of a phase-transition that identifies properties that makes security games hard to solve.
These new models have not only advanced the state of the art in computational game-theory, but have actually been successfully deployed in the real-world. My work represents a successful transition from game-theoretic advancements to real-world applications that are already in use, and it has opened exciting new avenues to greatly expand the reach of game theory. For instance, my algorithms are used in the IRIS system: IRIS has been in use by the Federal Air Marshals Service (FAMS) to schedule air marshals on board international commercial flights since October 2009.
Committee:
Milind Tambe (chair)
Bhaskar Krishnamachari
Matthew McCubbins (outside member)
Vincent Conitzer
Fernando Ordonez
Host: Lizsl de Leon
Location: 223
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: Nora Ayanian (MIT): Distributed Multirobot Coordination: From Specification to Provably Correct Execution
Thu, Apr 11, 2013 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Nora Ayanian, MIT
Talk Title: Distributed Multirobot Coordination: From Specification to Provably Correct Execution
Abstract: Using a group of robots in place of a single complex robot to accomplish a task has many benefits, including simplified system repair, less down time, and lower cost. Combining heterogeneous groups of these multi-robot systems allows addressing multiple subtasks in parallel, reducing the time it takes to address many problems, such as search and rescue, reconnaissance, and mine detection. These missions demand different roles for robots, necessitating a strategy for coordinated autonomy while respecting any constraints the environment may impose. Distributed computation of control policies for heterogeneous multirobot systems is particularly challenging because of inter-robot constraints such as communication maintenance and collision avoidance, the need to coordinate robots within groups, and the dynamics of individual robots.
I will present algorithms for synthesizing distributed globally convergent feedback policies for navigating groups of heterogeneous robots in known constrained environments. Provably correct by construction, these algorithms automatically and concurrently solve both the path planning and control synthesis subproblems by decomposing the space into cells and sequentially composing local feedback controllers. The approach is useful for many decentralized applications of multirobot systems including task allocation, navigation in formation, and human-robot interaction. Finally, I will extend the algorithm to partially known environments, where dynamic task reassignment allows the team to cope with unknown hazards in the environment while still providing guarantees on convergence and safety.
Biography: Nora Ayanian is a postdoctoral associate in the Computer Science and Artificial Intelligence Laboratory at MIT. She received a M.S. and Ph.D. in Mechanical Engineering at the University of Pennsylvania, in 2008 and 2011, respectively.
Host: Fei Sha
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
Shiri Chechik (Microsoft Research) Distance Oracles with Local Stretch
Fri, Apr 12, 2013 @ 12:00 PM - 01:30 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Shiri Chechik , Microsoft Research Silicon Valley
Talk Title: Distance Oracles with Local Stretch
Series: USC CS Theory
Abstract: A Distance Oracle is a succinct data structure that provides fast answers to distance queries between any two points.
Distance oracles are measured by several parameters: construction time (the running time of the algorithm to produce the data structure), size (the worst case size of the data structure), query complexity (the running time of the query algorithm, given two points), and stretch guarantee (the maximum ratio between the estimated distance returned by the distance oracle and the actual distance).
In this talk we will consider a more refined local stretch guarantee first suggested by Abraham, Bartal and Neiman [STOC 07]. Informally, we wish to obtain better stretch guarantees for nearby pairs. We would like the stretch bound to gradually improve as we query closer pairs of points.
We consider two notions of local stretch for distance oracles:
1. Strong local stretch provides stretch guarantees for any pair of nodes with better stretch guarantees for nearby pairs.
2. Weak local stretch provides stretch guarantees only between pair of nodes u and v, such that v is in the r-neighborhood of u (v is one of the rââ¬â¢th closest nodes to u), for some parameter r.
We will discuss these two notions and see efficient constructions for both these notions improving upon previous work.
Based on a joint work with Ittai Abraham
Biography: Shiri Chechik is Postdoctoral Researcher at Microsoft Research Silicon Valley. She recently completed her PhD under the supervision of Prof. David Peleg in the Weizmann Institute of Science in Israel.
She is interested in Theoretical Computer Science, with an emphasis on Design and Analysis of Algorithms for network problems.
Home page: http://research.microsoft.com/en-us/people/schechik/.
Host:
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
2013 Symposium on Game Theory and Human Behavior
.jpg)
Tue, Apr 16, 2013
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Noam Nisan (Computer Science, Hebrew U.) Duncan Watts (Microsoft Research) David Pennock (Microsoft Research) Asu Ozdaglar (EE & CS, MIT) Matthew Elliot (Economics, CalTech) Timothy Ketelaar (Psychology, NMSU) Alan Sanfrey (Psychology, U Arizona),
Talk Title: GTHB Annual Conference
Series: Game Theory and Human Behavior Group
Abstract: Full Event information to be announced closer to the date. Check back soon!
RSVP to Rishika Agarwal to attend at rishikaa@usc.edu
Speakers include:
Noam Nisan (Computer Science, Hebrew U.)
Duncan Watts (Microsoft Research)
David Pennock (Microsoft Research)
Asu Ozdaglar (EE & CS, MIT)
Matthew Elliot (Economics, CalTech)
Timothy Ketelaar (Psychology, NMSU)
Alan Sanfrey (Psychology, U Arizona)
Host: Milind Tambe
Location: USC University Club at King Stoops Hall
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
Nadia Heninger (Princeton): RSA in the real world
Tue, Apr 16, 2013 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Nadia Heninger, Princeton
Talk Title: RSA in the real world
Series: CS Colloquium
Abstract: I study computer security and applied cryptography using a theoretician's mathematical toolkit. Security vulnerability analysis can often be a painstaking and implementation-specific process. My approach uses cryptographic and algorithmic ideas to reason about the security of deployed systems, to question assumptions underlying the security of these systems, and to understand and model threats.
In this talk, I will use RSA, the world's most widely used public key cryptosystem, as a vehicle to explore the interaction between cryptographic algorithms and real-world usage:
- Discovering widespread catastrophic failures in the random number generators in network devices by computing the greatest common divisors of millions of RSA public keys collected in the wild.
- Reconstructing complete private keys using only a few bits of the private key revealed in the course of a side-channel attack.
In addition to their impact on security, many of the ideas arising in the course of this work have surprising connections across computer science, leading to, for example, new algorithms for decoding families of error-correcting codes, applications within theoretical cryptography, and practical privacy-enhancing technologies.
Biography: Nadia Heninger is a visiting researcher at Microsoft Research New England. Her research focuses on security, applied cryptography, and algorithms. She is best known for her work identifying widespread entropy problems in cryptographic keys on the Internet (2012 Usenix Security best paper award), and developing the "cold boot" attack against disk encryption systems (2008 Usenix Security best student paper award). In 2011-2012, she was an NSF Mathematical Sciences Postdoctoral Fellow at UC San Diego. She received her Ph.D. in computer science in 2011 from Princeton and a B.S. in electrical engineering and computer science in 2004 from UC Berkeley.
Host: Ramesh Govindan
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Distinguished Lecture: Gail Kaiser (Columbia): Testing 1... 2... 3...
Thu, Apr 18, 2013 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Gail Kaiser, Columbia
Talk Title: Testing 1... 2... 3...
Series: CS Distinguished Lectures
Abstract: Testing software systems is hard. Conventional software testing checks whether each output is correct for the set of test inputs. But for some software, it is not known what the correct output should be for some inputs. How can we construct and execute test cases that will find coding errors even when we do not know whether the output is correct? And for most software, the development-lab testing process can not cover all inputs and/or internal states that can arise after deployment. How can we construct and execute test cases that operate in the states that occur during user operation to continue to find coding errors without impacting the user? Finally, for some (most?) software, even with rigorous pre and post deployment testing, users will inevitably notice errors that were not detected by the developer's test cases. How can we construct and execute new test cases that reproduce these errors? This talk will present an overview of my lab's past decade and ongoing research on these hard testing problems.
Biography: Gail E. Kaiser is a Professor of Computer Science and the Director of the Programming Systems Laboratory in the Computer Science Department at Columbia University. She was named an NSF Presidential Young Investigator in Software Engineering and Software Systems in 1988, and has published over 150 refereed papers in a range of software areas. Prof. Kaiser's research interests include social software engineering, collaborative work, privacy and security, software reliability, self-managing systems, parallel and distributed systems, Web technologies, information management, and software development environments and tools. She has consulted or worked summers for courseware authoring, software process and networking startups, several defense contractors, the Software Engineering Institute, Bell Labs, IBM, Siemens, Sun and Telcordia. Her lab has been funded by NSF, NIH, DARPA, ONR, NASA, NYS Science & Technology Foundation, and numerous companies. Prof. Kaiser served on the editorial board of IEEE Internet Computing for many years, was a founding associate editor of ACM Transactions on Software Engineering and Methodology, chaired an ACM SIGSOFT Symposium on Foundations of Software Engineering, vice chaired three of the IEEE International Conference on Distributed Computing Systems, and serves frequently on conference program committees. She also served on the Committee of Examiners for the Educational Testing Service's Computer Science Advanced Test (the GRE CS test) for three years, and has chaired her department's doctoral program since 1997. Prof. Kaiser received her PhD and MS from CMU and her ScB from MIT.
Host: William GJ Halfond
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Distinguished Lecture: Lise Getoor (University of Maryland College Park): Statistical Relational Learning and Graph Identification
Mon, Apr 22, 2013 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Lise Getoor, University of Maryland College Park
Talk Title: Statistical Relational Learning and Graph Identification
Series: CS Distinguished Lectures
Abstract: Within the machine learning and data mining communities, there is a growing interest in learning structured models from input data that is itself structured, an area often referred to as statistical relational learning (SRL). In this talk, Iââ¬â¢ll give a brief overview of SRL, discuss its relation to graph analysis, extraction, and alignment, and its importance in the context of big data analytics. Iââ¬â¢ll then describe our recent work on "graph identification", the process of inferring a graph or network from observational data. Graph identification requires a combination of entity resolution (determining when two references refer to the same underlying entity), link prediction (inferring missing relationships in the data), and collective classification (inferring attribute values of the entities). This form of structured prediction allows us to infer missing information and correct mistakes -- a vital first step before further network analysis is performed. I will overview two approaches to graph identification: 1) coupled conditional classifiers (C^3), and 2) probabilistic soft logic (PSL). I will describe their mathematical foundations, learning and inference algorithms, and empirical evaluation, showing their power in terms of both accuracy and scalability. These methods support emerging information extraction and database techniques to realize the promise of extracting actionable knowledge from large-scale data in the wild. I will conclude by highlighting connections to privacy in social network data and other big data challenges.
Biography: Lise Getoor is an Associate Professor in the Computer Science Department at the University of Maryland, College Park and University of Maryland Institute for Advanced Computer Studies. Her research areas include machine learning, and reasoning under uncertainty; in addition she works in data management, visual analytics and social network analysis. She is a board member of the International Machine Learning Society, a former Machine Learning Journal Action Editor, Associate Editor for the ACM Transactions of Knowledge Discovery from Data, JAIR Associate Editor, and she has served on the AAAI Council. She was conference co-chair for ICML 2011, and has served on the PC of many conferences including the senior PC for AAAI, ICML, KDD, UAI and the PC of SIGMOD, VLDB, and WWW. She is a recipient of an NSF Career Award and was awarded a National Physical Sciences Consortium Fellowship. Her work has been funded by ARO, DARPA, IARPA, Google, IBM, LLNL, Microsoft, NGA, NSF, Yahoo! and others. She received her PhD from Stanford University, her Masterââ¬â¢s degree from University of California, Berkeley, and her undergraduate degree from University of California, Santa Barbara.
Host: Leana Golubchik
Location: Seaver Science Library (SSL) - 150
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
.jpg)