Conferences, Lectures, & Seminars
Events for September
-
AI Seminar
Fri, Sep 06, 2013 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Liang Huang, City University of New York (CUNY)
Talk Title: Scalable Training for Machine Translation Made Successful for the First Time
Abstract: While large-scale discriminative training has triumphed in many NLP problems, its definite success on machine translation has been largely elusive. Most recent efforts along this line are not scalable: they only train on the small dev set with an impoverished set of rather ââ¬Ådenseââ¬Â features. We instead present a very simple yet theoretically motivated approach by extending my recent framework of ââ¬Åviolation-fixing perceptronââ¬Â to the latent variable setting, and use forced decoding to compute the target derivations. Our method allows structured learning to scale, for the first time, to a large portion of the training data, which enables a rich set of sparse, lexicalized, and non-local features. Extensive experiments show very significant gains in BLEU (by at least +2.0) over MERT and PRO baselines with the help of over 20M sparse features.
Biography: Liang Huang is currently an Assistant Professor at the City University of New York (CUNY). He graduated in 2008 from Penn and has worked as a Research Scientist at Google and a Research Assistant Professor at USC/ISI. His work is mainly on the theoretical aspects (algorithms and formalisms) of computational linguistics, and related theoretical problems in machine learning. He has received a Best Paper Award at ACL 2008, several best paper nominations (ACL 2007, EMNLP 2008, and ACL 2010), two Google Faculty Research Awards (2010 and 2013), and a University Graduate Teaching Prize at Penn (2005).
Host: David Chiang
Location: Information Science Institute (ISI) - 6th floor conference room
Audiences: Everyone Is Invited
Contact: Kary LAU
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
NL Seminar-Jeon-Hyung Kang:"LA-CTR: A Limited Attention Collaborative Topic Regression for Social Media"
Fri, Sep 06, 2013 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Jeon-Hyung Kang, USC/ISI
Talk Title: "LA-CTR: A Limited Attention Collaborative Topic Regression for Social Media"
Series: Natural Language Seminar
Abstract: Abstract: Probabilistic models can learn usersââ¬â¢ preferences from the history of their item adoptions on a social media site, and in turn, recommend new items to users based on learned preferences. However, current models ignore psychological factors that play an important role in shaping online social behavior. One such factor is attention, the mechanism that integrates perceptual and cognitive features to select the items the user will consciously process and may eventually adopt. Recent research has shown that people have finite attention, which constrains their online interactions, and that they divide their limited attention non-uniformly over other people. We propose a collaborative topic regression model that incorporates limited, non-uniformly divided attention. We show that the proposed model is able to learn more accurate user preferences than state-of-art models, which do not take human cognitive factors into account. Specifically we analyze voting on news items on the social news aggregator and show that our model is better able to predict held out votes than alternate models. Our study demonstrates that psycho-socially motivated models are better able to describe and predict observed behavior than models which only consider latent social structure and content.
Biography: Home Page:http://isi.edu/integration/people/kang/
Host: Yang Gao
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 11th Flr Conf Rm # 689, Marina Del Rey
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: http://nlg.isi.edu/nl-seminar/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
AI SEMINAR - Miel Vander Sande
Fri, Sep 13, 2013 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Miel Vander Sande, Ghent University-iMinds, Belgium
Talk Title: The story behind Everything is Connected: Multimedia narration of automatically discovered paths in Linked Data
Abstract: Everything is Connected (http://everythingisconnected.be) is a Linked Data application for automatically generating a story between two concepts in the Web of Data, based on formally described links. A path between two concepts is obtained by browsing linked open datasets; the path is then enriched with multimedia presentation material for each node in order to obtain a full multimedia presentation of the found path. An efficient technique combining pre-processing and indexing of RDF datasets is used, which is able to find paths in a couple of seconds.
Biography: Miel Vander Sande is a researcher in computer science and engineering at the Multimedia Lab of Ghent University-iMinds, Belgium. In 2010, he graduated as M.Sc. in Industrial Science (specialisation on ICT) from University College of Ghent, Belgium. As from 2011, he was involved in several research projects involving Semantic Web technologies and Open Data. Miel Vander Sande's main interest and expertise are (linked open) data publishing (a.o. in the context of Open Knowledge Foundation), technological support for open data legislation, ontology mapping and data transformation. He is also member of the W3C Linked Data Platform working group. Together with his team, Miel developed the Everything is Connected (http://www.everythingisconnected.be) demo. This demo makes Linked Data tangible by finding paths between concepts using the DBPedia dataset. This resulted in a novel high performance approach for performing graph algorithms on generic datasets. Miel's demo paper was accepted at the International Semantic Web Conference 2012 in Boston, where it won the Best Demo Award. Currently, Miel is active in multiple Flemish government projects for creating read/write Open Data ecosystems. Additionally, he participates in a Flemish innovation project for the digitalization of book publishers and eBooks: Publishers of the future.
Host: Craig Knoblock, USC/ISI
Webcast: http://webcasterms1.isi.edu/mediasite/Viewer/?peid=45b27e6ddec34405a638789183ac85421dLocation: Information Science Institute (ISI) -
WebCast Link: http://webcasterms1.isi.edu/mediasite/Viewer/?peid=45b27e6ddec34405a638789183ac85421d
Audiences: Everyone Is Invited
Contact: Alma Nava / Information Sciences Institute
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
NL Seminar- Kevin Knight: "Some Potential NLP Thesis Topics & Other Fun Research Projects"
Fri, Sep 13, 2013 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Kevin Knight, USC/ISI
Talk Title: "Some Potential NLP Thesis Topics & Other Fun Research Projects"
Series: Natural Language Seminar
Abstract: I'll present a dozen interesting, potentially high-impact NLP research projects. I'd like to make this a very interactive session.
Biography: Home Page:http://www.isi.edu/~knight/
Host: Yang Gao
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: http://nlg.isi.edu/nl-seminar/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
AI SEMINAR
Thu, Sep 19, 2013 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Dr. Jens Lehmann, Head of AKSW/MOLE group, University of Leipzig
Talk Title: The LinkedGeoData and GeoKnow Projects
Series: AISeminar
Abstract: In the talk, Jens Lehmann will give an introduction to Geospatial Linked Data projects at his research group AKSW (http://aksw.org) at the University of Leipzig / Germany. The presentation will cover the GeoKnow (http://geoknow.eu) European Union project, which is a funded project with 7 partner organisations, as well as spatial data extraction in with 7 partner organisations, as well as spatial data extraction in LinkedGeoData (http://linkedgeodata.org), a large RDF knowledge base derived from OpenStreetMap.
Biography: Dr. Jens Lehmann (http://www.jens-lehmann.org) is a researcher at the University of Leipzig, at which he is co-leading the Agile Knowledge Engineering and Semantic Web Group. His research interests involve Semantic Web, machine learning and knowledge representation. He is founder, leader or contributor of several community research projects, including DL-Learner, DBpedia, LinkedGeoData and ORE. He works/worked in several funded projects, e.g. GeoKnow (EU STREP, coordinator), LOD2 (EU IP, work package lead), LATC (EU STREP, lead for University of Leipzig) and SoftWiki (BmBF). Dr. Jens Lehmann authored more than 40 articles in international journals and conferences cited more than 3000 times according to Google Scholar.
Host: Craig Knoblock, USC/ISI
Webcast: will not be webcastedLocation: Information Science Institute (ISI) -
WebCast Link: will not be webcasted
Audiences: Everyone Is Invited
Contact: Alma Nava / Information Sciences Institute
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
NL Seminar- Yang Feng: "A Markov Model of Machine Translation using Non-parametric Bayesian Inference (ACL 2013)'
Fri, Sep 20, 2013 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Yang Feng, USC/ISI
Talk Title: "A Markov Model of Machine Translation using Non-parametric Bayesian Inference (ACL 2013)"
Series: Natural Language Seminar
Abstract: Most modern machine translation systems use phrase pairs as translation units, allowing for accurate modeling of phrase-internal translation and reordering. However phrase-based approaches are much less able to model sentence level effects between different phrase-pairs. We propose a new model to address this imbalance, based on a word-based Markov model of translation which generates target translations left-to-right. Our model encodes word and phrase level phenomena by conditioning translation decisions on previous decisions and uses a hierarchical Pitman-Yor Process prior to provide dynamic adaptive smoothing. This mechanism implicitly supports not only traditional phrase pairs, but also gapping phrases which are non-consecutive in the source.
Biography: Yang Feng is a posdoc of the natural language group in USC/ISI. She got her ph.D degree from Institute of Computing Technology, Chinese Academy of Sciences. Her research interests are in all aspects of machine translation and machine learning focusing on graphical models and Bayesian inference.
Home Page:
http://www.isi.edu/~yangfeng
Host: Yang Gao
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: http://nlg.isi.edu/nl-seminar/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor. -
NL Seminar- Andrew S. Gordon: "Heider-Simmel Interactive Theater "
Fri, Sep 27, 2013 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Andrew S. Gordon, USC/ ICT
Talk Title: "Heider-Simmel Interactive Theater "
Series: Natural Language Seminar
Abstract: In a famous 1944 paper, psychologist Fritz Heider and his student Marianne Simmel described an experiment where undergraduates were shown a short animated film depicting the movement of geometric shapes. Asked to describe what happened in the film, these students produced narratives that described the behavior of these shapes in anthropomorphic terms, ascribing to them plans, goals, emotions, and social roles that accounted for their behavior. Fritz Heider later wrote his seminal book, The Psychology of Interpersonal Relations, which articulated the role of Commonsense Psychology in the interpretation of the behavior of other people. In this talk I'll discuss our recent efforts to model the reasoning of the students in Heider and Simmel's original experiment. I'll describe our vision of a "Heider-Simmel Interactive Theater," a software application where people can create their own short movies involving geometric shapes in the style of Heider and Simmel's original film, which are then interpreted by the computer to generate a textual narrative of the author's creation. Then I'll lay out the technical plan, which involves the integration of probabilistic graphical models, weighted abduction, data-driven text generation, logical formalizations of commonsense psychology, and game-based data collection from the public at large. Before coming to the talk, please sign up and play "Triangle Charades" at the following website: http://charades.ict.usc.edu
Biography: http://people.ict.usc.edu/~gordon/
Host: Yang Gao
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: http://nlg.isi.edu/nl-seminar/
This event is open to all eligible individuals. USC Viterbi operates all of its activities consistent with the University's Notice of Non-Discrimination. Eligibility is not determined based on race, sex, ethnicity, sexual orientation, or any other prohibited factor.
