Conferences, Lectures, & Seminars
Events for September
-
AI Seminar
Wed, Sep 07, 2016 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: David Kale, USC
Talk Title: Computational Phenotyping: Combining Big Data, Flexible Models, and Domain Knowledge
Abstract: In this talk, I will discuss the challenges and opportunities of applying machine learning to digital health data in the context of computational phenotyping. Phenotyping involves the development of algorithms to answer questions like, "Does this patient have diabetes?" and has a wide variety of applications: cohort construction for genomic studies, risk adjustment, quality improvement, and diagnosis. In recent years, researchers have moved away from algorithmic disease definitions based on clinical knowledge, which are expensive to develop and validate, and toward data-driven phenotypes based on the application of machine learning to large healthcare databases. I will provide an overview of phenotyping and its applications in medicine, discuss recent trends in the field, and present my recent work on phenotyping clinical time series with recurrent neural networks. I will also discuss ongoing work to develop methods that can exploit available data and domain knowledge to train data-driven models in the absence of ground truth training
Biography: Dave Kale is a fifth year PhD student in Computer Science and an Alfred E. Mann Innovation in Engineering Fellow at the University of Southern California. He is advised by Prof. Greg Ver Steeg at the USC Information Sciences Institute, a member of Aram Galstyan's lab at ISI, and an affiliate of Nigam Shah's lab at the Stanford Center for Biomedical Informatics Research. Dave co-founded the Machine Learning for Healthcare Conference (MLHC), the preeminent venue for research on machine learning applied to health. Dave holds a BS and MS from Stanford University
Host: Emilio Ferrara
Location: Information Science Institute (ISI) - 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-How we Cracked the Borg Cipher + First Steps Towards Deciphering from Images
Fri, Sep 09, 2016 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Nada Aldarrab, USC/ISI
Talk Title: How we Cracked the Borg Cipher + First Steps Towards Deciphering from Images
Series: Natural Language Seminar
Abstract: European libraries are filled with undeciphered historical manuscripts from the 16th-18th centuries. These documents are enciphered with classical methods, which puts their contents out of the reach of historians who are interested in the history of that era. In this talk, we show how we automatically cracked a 400-page book from the 17th century. We also describe a system aimed at deciphering from camera-phone images. We show initial results for different ciphers.
Biography: Nada is a graduate student at USC, working on her thesis under the supervision of Prof. Kevin Knight. She is currently working on the decipherment of historical documents (joint project with Uppsala University, Sweden). Her research interests include natural language processing, machine learning, decipherment and machine translation.
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. -
AI SEMINAR
Thu, Sep 15, 2016 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Rudi Studer, Institutes AIFB/KSRI, Karlsruhe Institute of Technology (KIT), Germany & FZI Research Center for Information Technology at KIT, Karlsruhe, Germany
Talk Title: Flexible Management of Event Processing Applications for the (Industrial) Internet of Things
Series: AI Seminar
Abstract: Gathering and processing events from cyber-physical systems provides users with the opportunity to continuously be aware of current performance indicators and potentially upcoming issues as well as to optimize production and maintenance processes. In this context, Event Processing has become an established technology to process high-frequency event streams in real-time while providing capabilities to detect event patterns based on spatial, temporal or causal relationships.
However, although event processing applications are often highly dynamic in regard to oftentimes changing requirements of observed situations as well as frequent syntactic and semantic changes of incoming sensor data, current technologies still suffer from high technical complexity making the development of real-time applications a time-consuming task due to slow development cycles.
In this talk, we discuss methods and tools supporting the management of event processing applications. We present a lightweight, semantics-based model to describe event sources such as sensors, event processing agents and consumers and an approach that enables application specialists to define and execute event processing pipelines in a self-service manner. The approach is illustrated based on two IoT scenarios: integrated monitoring of manufacturing processes and disruption management in supply chains. In the beginning, selected research activities at Institutes AIFB and KSRI at KIT as well as at FZI will be outlined.
Biography: Bio:
Rudi Studer is Full Professor in Applied Informatics at the Karlsruhe Institute of Technology (KIT), Institute AIFB. In addition, he is director at the Karlsruhe Service Research Institute (KSRI) as well as at the FZI Research Center for Information Technology at KIT. His research interests include knowledge management, Semantic Web technologies and applications, data and text mining, Big Data and Service Science.
He obtained a Diploma in Computer Science at the University of Stuttgart in 1975. In 1982 he was awarded a Doctor's degree in Informatics at the University of Stuttgart, and in 1985 he obtained his Habilitation in Informatics at the University of Stuttgart. From 1985 to 1989 he was project leader and manager at the Scientific Center of IBM Germany.
He is involved in various national and international research projects, among others the EU projects XLime (crossLingual crossMedia Knowledge Extraction) and iVision (Immersive Semantics-based Virtual Environments for the Design and Validation of Human-centered Aircraft Cockpits).
Rudi Studer is former president of the Semantic Web Science Association (SWSA) and former Editor-in-chief of the Journal of Web Semantics: Science, Services and Agents on the World Wide Web. He is a Semantic Technologies Institute (STI) International Fellow.
Host: Craig Knoblock
Webcast: http://webcastermshd.isi.edu/Mediasite/Play/33a9588370b74466825f95063e27108e1dLocation: Information Science Institute (ISI) - 1135 - 11th fl Large CR
WebCast Link: http://webcastermshd.isi.edu/Mediasite/Play/33a9588370b74466825f95063e27108e1d
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-Efficient Exploration for Dialog Policy Learning with BBQ Networks & Replay Buffer Spiking
Fri, Sep 16, 2016 @ 01:30 PM - 02:30 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Zachary Lipton, UCSD
Talk Title: Efficient Exploration for Dialog Policy Learning with BBQ Networks & Replay Buffer Spiking
Series: Natural Language Seminar
Abstract: When rewards are sparse and efficient exploration essential, deep Q learning with e greedy exploration tends to fail. This poses problems for otherwise promising domains such as task oriented dialog systems, where the primary reward signal, indicating successful completion, typically occurs only at the end of each episode but depends on the entire sequence of utterances. A poor agent encounters such successful dialogs rarely, and a random agent may never stumble upon a successful outcome in reasonable time. We present two techniques that significantly improve the efficiency of exploration for deep Q learning agents in dialog systems. First, we demonstrate that exploration by Thompson sampling, using Monte Carlo samples from a Bayes by Backprop neural network, yields marked improvement over standard DQNs with Boltzmann or e greedy exploration. Second, we show that spiking the replay buffer with a small number of successes, as are easy to harvest for dialog tasks, can make Q learning feasible when it might otherwise fail catastrophically.
Biography: I am a graduate student in the Artificial Intelligence Group at the University of California, San Diego on leave for two quarters at Microsoft Research Redmond. I work on machine learning, focusing on deep learning methods and applications. In particular, I work on modeling sequential data with recurrent neural networks and sequential decision-making processes with deep reinforcement learning. I'm especially interested in research impacting medicine and natural language processing. Recently, in Learning to Diagnose with LSTM RNNs, we trained LSTM RNNs to accurately predict patient diagnoses using only lightly processed time series of sensor readings in the pediatric ICU. Before coming to UCSD, I completed a Bachelor of Arts with a joint major in Mathematics and Economics at Columbia University. Then, I worked in New York City as a jazz musician. I have interned with Amazon's Core Machine Learning team and Microsoft Research's Deep Learning Team.
Host: Xing Shi and Kevin Knight
More Info: http://nlg.isi.edu/nl-seminar/
Location: Information Science Institute (ISI) - 6th Floor -CR # 689; ISI-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-DUOLINGO: IMPROVING LANGUAGE LEARNING AND ASSESSMENT WITH DATA
Mon, Sep 19, 2016 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Burr Settles, Duolingo
Talk Title: DUOLINGO: IMPROVING LANGUAGE LEARNING AND ASSESSMENT WITH DATA
Series: Natural Language Seminar
Abstract: Duolingo is a language education platform with more than 150 million students worldwide. Our flagship learning app is the #1 way to learn a language online, and is the most-downloaded education app for both Android and iOS devices. It is also completely free. In this talk, I will describe the Duolingo system and several empirical projects, which mix machine learning with computational linguistics and psychometrics to improve learning, engagement, and even language proficiency assessment through our products.
Biography: Burr Settles is a scientist, engineer, and head of research at Duolingo: the most widely used education application in the world, teaching 20 languages to more than 150 million users worldwide. He is also the principal developer of the Duolingo English Test: a computer-adaptive proficiency exam that aims to disrupt and democratize the global certification marketplace through highly accessible mobile technology. Before joining Duolingo, he earned a PhD in computer sciences at University of Wisconsin-Madison, and then worked as a postdoctoral research scientist at Carnegie Mellon University, where his work has spanned machine learning, natural language processing, and computational social science. His 2012 book Active Learning is now the standard text on learning algorithms that are adaptive, curious, or exploratory (if you will). Burr gets around by bike and (among other things) plays guitar in the pop band delicious pastries.
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. -
NL Seminar-POETRY AT THE METAPHORICAL INTERSECTION
Mon, Sep 26, 2016 @ 03:00 PM - 04:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Andrea Gagliano , (UC Berkeley)
Talk Title: Poetry at the Metaphorical Intersection
Series: Natural Language Seminar
Abstract: This talk will discuss a technique to create figurative relationships using Mikolov et als word vectors. Drawing on existing work on figurative language, we start with a pair of words and use the intersection of word vector similarity sets to blend the distinct semantic spaces of the two words. We conduct preliminary quantitative and qualitative observations to compare the use of this novel intersection method with the standard word vector addition method for the purpose of supporting the generation of figurative language. To showcase this technique, we use it to write computer generated sonnets.
Biography: Andrea Gagliano is a masters student at UC Berkeley's School of Information and the Berkeley Center for New Media. Her research explores the use of computation for creativity - both tools to support creative practices and generation of creative works. Recently, she has been focusing in the field of natural language processing by working on poetry and metaphor generation.
Previously, Andrea received her BS in Mathematics and BA in Business Administration from the University of Washington in 2013. During her studies, she spent time with the Creative Writing department studying poetry.
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.
