Select a calendar:
Filter October Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Conferences, Lectures, & Seminars
Events for October
-
CS Colloquium: Adrian Nistor (Chapman University) - Detecting and Repairing Performance Bugs using Execution and Code Patterns
Thu, Oct 09, 2014 @ 11:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Adrian Nistor, Chapman University
Talk Title: Detecting and Repairing Performance Bugs using Execution and Code Patterns
Series: CS Colloquium
Abstract: In this talk we will discuss Caramel, Toddler, and SunCat, three novel techniques for automatically detecting and repairing performance bugs. Unlike profilers, which focus on methods that take a long time to execute, Caramel, Toddler, and SunCat focus on code and execution patterns that are highly indicative of common programming mistakes affecting performance. The additional information provided by these patterns enable Caramel, Toddler, and SunCat to have better results---more automation, fewer false negatives, fewer false
positives, automated repair---than profilers for the bugs Caramel, Toddler, and SunCat are designed to find. Caramel, Toddler, and SunCat employ novel dynamic and static analyses. Caramel, Toddler, and SunCat found previously unknown performance bugs in widely used Java, C/C++, and C# applications, including in mobile applications.
Biography: Adrian Nistor started as an Assistant Professor at Chapman University in Fall 2014. He received his Ph.D from the Computer Science Department at the University of Illinois at Urbana-Champaign in May 2014. His research interests are in software engineering, with a focus on detecting, repairing, and preventing bugs in real-world applications. His projects investigate performance bugs and concurrency bugs. His techniques found more than 150 previously unknown bugs in widely used software, e.g., PARSEC, GCC, Google Chrome, Mozilla, MySQL, Ant, Google Core Libraries, Lucene, Tomcat, JUnit, JMeter, Log4J, etc. More than 100 of these bugs are already fixed by developers. His research includes empirical and analytical work, static and dynamic techniques, hardware-assisted and software-only solutions, and bugs from various application types---client, server, mobile, and scientific applications.
Host: GJ Halfond
Location: Henry Salvatori Computer Science Center (SAL) - 101
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium - USC Student Series: George Konstantinidis, Leandro Soriano Marcolino
Tue, Oct 14, 2014 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: George Konstantinidis, Leandro Soriano Marcolino, USC
Talk Title: Scalable Data Integration under Constraints, Agents Vote for the Environment: Designing Energy-Efficient Architecture
Series: Student Seminar Series
Abstract: Saving energy is a major concern nowadays. Hence, it is fundamental to design and construct buildings that are energy efficient. It is known that the early stage of architectural design has a significant impact on this matter. However, it is very complex to create designs that are optimally energy efficient, and at the same time balance and meet other essential design criteria such as economics, space, and safety. One state of the art approach is to create parametric designs, and use a genetic algorithm to optimize across complexly coupled objectives. In this work we further improve this method, by aggregating the solutions of multiple agents. We evaluate our approach across three design case studies of increasing complexity, and show that a team of agents are able to provide one order of magnitude higher number of 1st ranked solutions in the Pareto frontier. Therefore, our approach provides the designers with a higher number of optimized solutions to choose from, that they can further subjectively evaluate, thus leading to better and highly energy efficient building designs.
We witness an explosion of available data in all areas of human activity, from large scientific experiments, to medical data, to distributed sensors, to social media. Integrating data from disparate sources can lead to novel insights across scientific, industrial, and governmental domains. This integration is achieved by either creating a data warehouse, that is, by copying/transforming the data to a centralized site under a single schema for subsequent analysis (data exchange), or by leaving the data at their original sources and querying the data at analysis time ((virtual) data integration), making use of mappings or views between the source and the global schemas. In this work, we focus in scalable data integration and data exchange under constraints or dependencies (or ontologies). In both these problems we make use of the chase algorithm, a forward-chaining reasoning algorithm and the main tool to reason with dependencies. Our first contribution is to introduce the frugal chase, which produces smaller solutions than the standard chase, still remaining polynomial in data complexity. Our second contribution is to use the frugal chase to scale up virtual data integration, aka query answering using views, under constraints in the language of LAV-weakly-acyclic dependencies, a useful language capturing the W3C Recommendation RDF/S. The latter problem can be reduced to query rewriting using views without constraints by chasing the source mappings using the constraints. We construct a compact graph-based representation of the mappings and the constraints and develop an efficient algorithm to run the frugal chase on this representation. We show experimentally that our approach scales to larger problems, outperfomring the standard chase algorithm by close to two orders of magnitude and improving online data integration time by a factor of 3.
Biography: George Konstantinidis is a Ph.D. Candidate in the Computer Science Department at the University of Southern California (USC) and a Research Assistant at the Information Sciences Institute (ISI) at USC. He studied Computer Science at the University of Crete, Greece, and holds a M.Sc. degree in Computer Science from the University of Crete and the Foundation for Research and TechnologyHellas (FORTH). His research interests lie in the intersection of Databases and Artificial Intelligence, Knowledge Representation, Information and Data Integration and the Semantic Web, and in particular OntologyBased Data Answering, Integration and Evolution. He enjoys combining novel theory and practical implementations. He has been a reviewer for IJCAI and TKDE, and has published papers both in A.I. (e.g., ECAI, KAIS) and in Databases (e.g., SIGMOD, VLDB).
Leandro Soriano Marcolino is a 4th year PhD student at University of Southern California (USC). He is advised by Milind Tambe. Previously he was awarded the Monbukagakusho scholarship and obtained a M.A. in Systems Information Sciences in Future University Hakodate, Japan. His research work performed during his master's studies was a best paper nominee at AAMAS 2011. He has been researching since very early as an undergraduate student at Universidade Federal de Minas Gerais, Brazil, and was able to publish many papers before even entering graduate school. His research is mainly about teamwork and cooperation, and he has published on the topic in a variety of different domains such as swarm robotics, computer Go, and building design. He has published in several prestigious conferences in AI and robotics, such as AAAI, AAMAS, IJCAI, ICRA and IROS.
Host: CS Department
Location: Henry Salvatori Computer Science Center (SAL) - 101
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
RASC Seminar: Jeremy L. Wyatt (University of Birmingham) - Robots in Our World: Uncertain, Incomplete and Unfamiliar
Thu, Oct 23, 2014 @ 11:00 AM - 12:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Jeremy L. Wyatt, University of Birmingham
Talk Title: Robots in Our World: Uncertain, Incomplete and Unfamiliar
Series: RASC Seminar Series
Abstract: To make transfer to applications in everyday domains robots require the ability to cope with novelty, incomplete information and uncertainty. In this talk I will describe a line of work carried out over ten years that provides methods to tackle this. In particular I will focus on two problems: object search and manipulation. Both require the ability to reason about open or novel worlds. The results are demonstrated in a variety of robot systems: in particular the Dora and Boris robots. Dora is one of the first mobile robots able to plan in open worlds, using the notion of assumptions. Dora also uniquely attempts to explain and then verify explanations in the face of failure. Boris is a robot system for manipulation that can grasp novel objects, and if there is time I will also describe algorithms we are developing for Boris that allow active gathering of information to support manipulation.
Biography: Jeremy L Wyatt is Professor of Robotics and Artificial Intelligence at the University of Birmingham. He gained his PhD from Edinburgh in 1996. He has published more than 80 papers, been the recipient of two best paper awards, and has led a variety of international robotics projects. He is interested in particular in robot planning and learning.
Host: Stefan Schaal
Location: Ronald Tutor Hall of Engineering (RTH) - 406
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: Bilge Mutlu (University of Wisconsin-Madison) - Human-Centered Methods and Principles for Designing Robotic Products
Tue, Oct 28, 2014 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Bilge Mutlu , University of Wisconsin-Madison
Talk Title: Human-Centered Methods and Principles for Designing Robotic Products
Series: CS Colloquium
Abstract: Robotic products constitute an emerging family of technologies that holds tremendous promise for everyday use. This promise also presents challenges for designers: the interactions they afford can be far more complex than those with conventional products, and designing for these interactions introduces many new questions. For instance, how can we design a product that follows human social norms? What is the design space for such a product? How can we empower designers to tackle such design problems? In this talk, I will present my group's work on building human-centered tools, methods, and knowledge to enable the design of robotic products. In particular, I will describe the development of novel tools and methods that support complex design tasks across the key stages of the design process, including analysis, synthesis, and evaluation, and an exploration into the design space for robotic products across different platforms, including social, assistive, and telepresence robots.
Biography: Bilge Mutlu is an assistant professor of computer science, psychology, and industrial engineering at the University of Wisconsin-Madison. He received his Ph.D. degree from Carnegie Mellon University's Human-Computer Interaction Institute in 2009. His background combines training in interaction design, human-computer interaction, and robotics with industry experience in product design and development. Dr. Mutlu is a former Fulbright Scholar and the recipient of the NSF CAREER award and several paper awards and nominations, including HRI 2008, HRI 2009, HRI 2011, UbiComp 2013, IVA 2013, RSS 2013, and HRI 2014. His research has been covered by national and international press including the NewScientist, MIT Technology Review, Discovery News, Science Nation, and Voice of America. He has served in the Steering Committee of the HRI Conference and the Editorial Board of IEEE Transactions on Affective Computing, co-chairing the Program Committees for HRI 2015 and ICSR 2011 and the Program Sub-committees on Design for CHI 2013 and CHI 2014.
Host: Maja Mataric
More Information: Mutlu-Photo.jpg
Location: Henry Salvatori Computer Science Center (SAL) - 101
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
Day-After-HotNets Mini-Workshop at USC Computer Science
Wed, Oct 29, 2014 @ 10:00 AM - 03:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Multiple, See Event Details
Talk Title: Day-After-HotNets Mini-Workshop at USC Computer Science
Series: CS Colloquium
Abstract: We hope we will see you at HotNets at the University of Southern California on October 27-28! The HotNets General Chairs (John Heidemann and Ethan Katz-Bassett) and the USC Networked Systems Lab (Ramesh Govindan, Minlan Yu, Wyatt Lloyd, and Ethan Katz-Bassett) would like to invite you to stick around for part of the following day October 29 to see various talks on related work and attend other talks as part of a mini-workshop, in our group's newly renovated lab on the USC campus a short walk from the HotNets venue.
Biography: David Oran: 10:00-10:25
Brighten Godfrey: 10:25-10:50
Fahad Dogar: 10:50-11:15
âImproving Response Times of Data Center Applicationsâ
Luigi Rizzo: 11:15-11:40
lunch: 11:40-12:40
Nina Taft: 12:40 - 1:05
âA New Frontier for Privacy: Data Mining on Encrypted Dataâ
Te-Yuan (TY) Huang: 1:05-1:30
âA Buffer-Based Approach to Rate Adaptation: Evidence from a Large Video Streaming Serviceâ
Host: Ethan Katz-Bassett, Minlan Yu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 132
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: Brian Scassellati (Yale University) - Building Models of Self and Task
Wed, Oct 29, 2014 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Brian Scassellati, Yale University
Talk Title: Building Models of Self and Task
Series: CS Colloquium
Abstract: This talk is an amalgamation of two topics that came out of research on building socially collaborative systems that focus on building richer representations of both robots and the tasks that they engage in. First, I will discuss methods for building self-trained models of a robot's own kinematic structure and sensory systems. Second, I will describe on-going efforts to automatically learn hierarchical representations of task structure from observations. These two topics, taken together, present a novel viewpoint of how we can restructure the way in which we view the division between built-in representations and learned methods.
Biography: Brian Scassellati is a Professor of Computer Science, Cognitive Science, and Mechanical Engineering at Yale University and Director of the NSF Expedition on Socially Assistive Robotics. His research focuses on building embodied computational models of human social behavior, especially the developmental progression of early social skills. Using computational modeling and socially interactive robots, his research evaluates models of how infants acquire social skills and assists in the diagnosis and quantification of disorders of social development (such as autism).
Host: Maja Mataric
Location: Seeley G. Mudd Building (SGM) - 101
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
-
CS Colloquium: Chris J. C. Burges (Microsoft Research )
Thu, Oct 30, 2014 @ 03:30 PM - 05:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Chris J. C. Burges , Microsoft Research
Talk Title: From Machine Learning to Machine Comprehension? Plus, two challenge datasets, and a family relations model.
Series: CS Colloquium
Abstract: Will current Machine Learning approaches be sufficient to solve the problem of the machine comprehension of text? First I will explain why, despite its recent successes in attacking several basic problems in AI, I think that the answer is "probably not". Then I will describe two datasets that we recently created to spur research on this problem. Finally I will describe recent work on modeling relations between relations, as instantiated by a model of human family relations.
Biography: After a checkered past involving first theoretical physics and then systems engineering at Bell Labs, Chris Burges saw a cool demo of a neural network reading handwritten digits: this triggered his long descent into machine learning. He has worked on handwriting and machine print recognition (he worked on a system now used to read millions of checks daily, and on Zip code and handwritten address recognition for the USPS), support vector machines, audio fingerprinting (his work is currently used in XBox and Windows Media Player to identify music), speaker verification, and information retrieval. His ranking algorithm is currently used by Bing for web search. Chris manages the Machine Learning and Intelligence group at Microsoft Research. He was program co-chair of Neural Information Processing Systems 2012 and general co-chair of NIPS 2013. His main current research interest is on modeling meaning in language.
Host: Fei Sha
Location: Henry Salvatori Computer Science Center (SAL) - 101
Audiences: Everyone Is Invited
Contact: Assistant to CS chair
