Conferences, Lectures, & Seminars
Events for December
-
AI Seminar
Fri, Dec 02, 2016 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Thomas Lemberger, EMBO, Heidelberg, Germany
Talk Title: SourceData: a semantic platform to make published data and figures discoverable
Abstract: In scientific publications, data are visually depicted in figures or tables. The original data behind the figures the source data however are almost never available in a structured format that would make them findable and reusable. To address this issue, SourceData (http://sourcedata.embo.org) has built a suite of tools to capture the structure of published research data and to make published research papers discoverable based solely on their data content. SourceData converts the narrative descriptions provided in figure legends into standardized, machine readable metadata. Each biological component in a figure is consistently identified via links to established public databases of biological terms. The experimental design is furthermore captured in a structured format by classifying the role of each component. Computer assisted manual identification and classification of biological entities is performed with a web-based curation tool. A separate interface allows authors to verify the accuracy of curated information. In a pilot project, the SourceData team has processed over 15,000 experiments from papers across 23 journals. The resulting web of connected data can be browsed through the SmartFigure application (http://smartfigures.net), which displays data in the context of related figures published in other papers and enables users to easily navigate between them. Users can also use the SourceData search engine to directly retrieve data based on the design of an experiment. SourceData searches the structure of the data rather than relying on keyword indexing, thus avoiding potentially subjective interpretation of results provided in the text.
Biography: Thomas Lemberger is Deputy Head of Scientific Publications at EMBO (embo.org) in Heidelberg, Germany, Chief Editor of the open access journal Molecular Systems Biology (msb.embopress.org) and Project Leader of the SourceData project (sourcedata.embo.org). Trained as a molecular biologist, Thomas earned his PhD at the University of Lausanne, Switzerland, where he studied hormonal regulation of gene expression by nuclear receptors. For his postdoctoral research, he moved to Heidelberg, Germany, where his research focused on the regulation of transcription in the brain. He joined EMBO as scientific editor in 2005 and assumed the editorial oversight of Molecular Systems Biology since launch of the journal. He has recently initiated the SourceData project to build an open platform that makes scientific publications discoverable based on their data content.
Host: Gully Burns
Location: 11th floor large 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 PROCEDURAL LANGUAGE AND KNOWLEDGE
Fri, Dec 02, 2016 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Yejin Choi, University of Washington
Talk Title: PROCEDURAL LANGUAGE AND KNOWLEDGE
Series: Natural Language Seminar
Abstract: Various types of how to knowledge are encoded in natural language instructions: from setting up a tent, to preparing a dish for dinner, and to executing biology lab experiments. These types of instructions are based on procedural language, which poses unique challenges. For example, verbal arguments are commonly elided when they can be inferred from context, e.g.,bake for 30 minutes, not specifying bake what and where. Entities frequently merge and split, e.g.,vinegar and oil merging into dressing, creating challenges to reference resolution. And disambiguation often requires world knowledge, e.g., the implicit location argument of stir frying is on stove. In this talk, I will present our recent approaches to interpreting and composing cooking recipes that aim to address these challenges. In the first part of the talk, I will present an unsupervised approach to interpreting recipes as action graphs, which define what actions should be performed on which objects and in what order. Our work demonstrates that it is possible to recover action graphs without having access to gold labels, virtual environments or simulations. The key insight is to rely on the redundancy across different variations of similar instructions that provides the learning bias to infer various types of background knowledge, such as the typical sequence of actions applied to an ingredient, or how a combination of ingredients e.g., flour, milk, eggs becomes a new entity e.g, wet mixture . In the second part of the talk, I will present an approach to composing new recipes given a target dish name and a set of ingredients. The key challenge is to maintain global coherence while generating a goal-oriented text. We propose a Neural Checklist Model that attains global coherence by storing and updating a checklist of the agenda e.g., an ingredient list with paired attention mechanisms for tracking what has been already mentioned and what needs to be yet introduced. This model also achieves strong performance on dialogue system response generation. I will conclude the talk by discussing the challenges in modeling procedural language and acquiring the necessary background knowledge, pointing to avenues for future research.
Biography: Yejin Choi is an assistant professor at the Computer Science & Engineering Department of University of Washington. Her recent research focuses on language grounding, integrating language and vision, and modeling nonliteral meaning in text. She was among the IEEEs AI Top 10 to Watch in 2015 and a co-recipient of the Marr Prize at ICCV 2013. Her work on detecting deceptive reviews, predicting the literary success, and learning to interpret connotation has been featured by numerous media outlets including NBC News for New York, NPR Radio, New York Times, and Bloomberg Business Week. She received her Ph.D. in Computer Science at Cornell University.
Host: Xing Shi and Kevin Knight
More Info: http://nlg.isi.edu/nl-seminar/
Location: 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-Multimodal Machine Comprehension: Tasks and Approaches
Fri, Dec 09, 2016 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Radi Soricut, Google
Talk Title: Multimodal Machine Comprehension: Tasks and Approaches
Series: Natural Language Seminar
Abstract: The ability of computer models to achieve genuine understanding of information as presented to humans (text, images, etc) is a long-standing goal of Artificial Intelligence. Along the way towards this goal, the research community has proposed solving tasks such as machine reading comprehension and computer image understanding. In this talk, we introduce two new tasks that can help us move closer to the goal. First, we present a multi-choice reading comprehension task, for which the goal is to understand a text passage and choose the correct summarizing sentence from among several options. Second, we present a multi-modal understanding task, posed as a combined vision-language comprehension challenge: identifying the most suitable text describing a visual scene, given several similar options. We present several baseline and competitive learning approaches based on neural network architectures, illustrating the utility of the proposed tasks in advancing both image and language comprehension. We also present human evaluation results, which inform a performance upper-bound on these tasks, and quantify the remaining gap between computer systems and human performance (spoiler alert: we are not there yet).
Biography: Radu Soricut is a Staff Research Scientist in the Research and Machine Intelligence group at Google. Radu has a PhD in Computer Science from University of Southern California, and has been with Google since 2012. His main areas of interest are natural language understanding, multilingual processing, natural language generation (from multimodal inputs), and general machine learning techniques for solving these problems. Radu has published extensively in these areas in top-tier peer-reviewed conferences and journals, and has won the Best Paper Award at the North American Association for Computational Linguistics Conference (NAACL) in 2015. Radu's current project looks at bridging natural language understanding and generation using neural techniques, in the context of Google's focus on making natural language an effective way of interacting with the world and the technology around us.
Host: Xing Shi and Kevin Knight
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
Fri, Dec 16, 2016 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Mason Porter, UCLA
Talk Title: Multilayer Networks
Series: AI Seminar
Abstract: Networks arise pervasively in biology, physics, technology, social science, and myriad other areas. Traditionally, a network is modeled as a graph and consists of a time-independent collection of entities (the nodes) that interact with each other via a single type of edge. However, most networks include multiple types of connections (which could represent, for example, different modes of transportation), multiple subsystems, and nodes and/or edges that change in time. The study of "multilayer networks", which is perhaps the most popular area of
network science, allows one to investigate networks with such complexities. In this talk, I'll give an introduction to multilayer networks and their applications.
Biography: Mason Porter earned a B.S. in applied mathematics from Caltech in 1998 and a Ph.D. from the Center for Applied Mathematics from Cornell University in 2002. He was a postdoc at Georgia Tech (math), Mathematical Sciences Research Institute, and Caltech (physics) before joining the faculty of the Mathematical Institute at University of Oxford in 2007. He was named Professor of Nonlinear and Complex Systems in 2014. A few months ago, he took up a position as Professor of Mathematics at UCLA. Porter is known for the diversity and interdisciplinarity of his research (and for his sharp wit). In networks and complex systems, Porter has contributed to myriad topics, including community structure in networks, core--periphery structure, social contagions, political networks, granular force networks, multilayer networks, temporal networks, and navigation in transportation systems. Other subjects he has studied include granular crystals, Bose--Einstein condensates, nonlinear optics, numerical evaluation of hypergeometric functions, quantum chaos, and synchronization of cows. Porter's awards include the 2014 Erd\H{o}s--R\'{e}nyi Prize in network science, a Whitehead Prize (London Mathematical Society) in 2015, the Young Scientist Award for Socio- and Econophysics (German Physical Society) in 2016, and teaching awards from University of Oxford in recognition of his lecturing and student mentorship. Porter was named a Fellow of the American Physical Society in October 2016.
Host: Emilio Ferrara
Webcast: http://webcastermshd.isi.edu/Mediasite/Play/ef4957a6864d4e1db06e15cba71b9b021dLocation: Information Science Institute (ISI) - 1135 - 11th fl Large CR
WebCast Link: http://webcastermshd.isi.edu/Mediasite/Play/ef4957a6864d4e1db06e15cba71b9b021d
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. -
AI Seminar-STREAM DATA MINING: A BIG DATA PERSPECTIVE
Tue, Dec 20, 2016 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Latifur Khan, University of Texas at Dallas
Talk Title: STREAM DATA MINING: A BIG DATA PERSPECTIVE
Series: Artificial Intelligence Seminar
Abstract: Data streams are continuous flows of data. Examples of data streams include network traffic, sensor data, call center records and so on. Data streams demonstrate several unique properties that together conform to the characteristics of big data (i.e., volume, velocity, variety and veracity) and add challenges to data stream mining. In this talk we will present an organized picture on how to handle various data mining techniques in data streams. Most existing data stream classification techniques ignore one important aspect of stream data: arrival of a novel class. We address this issue and propose a data stream classification technique that integrates a novel class detection mechanism into traditional classifiers, enabling automatic detection of novel classes before the true labels of the novel class instances arrive. Novel class detection problem becomes more challenging in the presence of concept-drift, when the underlying data distributions evolve in streams. In this talk we will show how to make fast and correct classification decisions under this constraint with limited labeled training data and apply them to real benchmark data. In addition, we will present a number of stream classification applications such as adaptive malicious code detection, website fingerprinting, evolving insider threat detection and textual stream classification.
This research was funded in part by NSF, NASA, Air Force Office of Scientific Research (AFOSR), IBM and Raytheon.
Biography: Dr. Latifur Khan is currently a full Professor (tenured) in the Computer Science department at the University of Texas at Dallas, USA where he has been teaching and conducting research since September 2000. He received his Ph.D. and M.S. degrees in Computer Science from the University of Southern California in August of 2000, and December of 1996 respectively. Dr. Khan is an ACM Distinguished Scientist. He has received prestigious awards including the IEEE Technical Achievement Award for Intelligence and Security Informatics.
Dr. Khan has published over 200 papers in prestigious journals, and in peer reviewed conference proceedings. Currently, his research area focuses on big data management and analytics, data mining, complex data management including geo-spatial data and multimedia data.
Host: Jose Luis Ambite
More Info: www.utdallas.edu/~lkhan/
Webcast: http://webcastermshd.isi.edu/Mediasite/Play/9ca403ad75674c839ec10e1c0ac9aee71dLocation: Information Science Institute (ISI) - 11th Flr Conf Rm # 1135, Marina Del Rey
WebCast Link: http://webcastermshd.isi.edu/Mediasite/Play/9ca403ad75674c839ec10e1c0ac9aee71d
Audiences: Everyone Is Invited
Contact: Peter Zamar
Event Link: www.utdallas.edu/~lkhan/
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.
