Select a calendar:
Filter August Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Conferences, Lectures, & Seminars
Events for August
-
NL SEMINAR Improving Dialogue Agents with a Social Dimension and Understanding Monolingual Pre-Training for Bilingual Models
Thu, Aug 13, 2020 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Naitian Zhou and Omar Shaikh , ISI interns
Talk Title: Improving Dialogue Agents with a Social Dimension and Understanding Monolingual Pre-Training for Bilingual Models
Abstract: Naitian Zhou
TITLE Improving Dialogue Agents with a Social Dimension
The dialogue problem is challenging because a proper response must be conditioned on many different factors: knowledge about the language, knowledge about the world, knowledge about self, and knowledge about the speaker, to name a few. Prior research has focused on language modeling and "persona" modeling, encoding facts about the dialogue agent. In this project, I try to focus on an alternative dimension to dialogue: how do we condition the way we converse, based on our understanding of ourselves and our social relationship with our dialogue partner?
Omar Shaikh
TITLE Understanding Monolingual Pre Training for Bilingual Models
Monolingual embeddings from models like BERT are known to help on a variety of downstream tasks in a straightforward way. Usually, these embeddings are plug and play initializing models with BERT embeddings or using them as input representations result in increased model performance. However, supervised NMT tasks do not appear to benefit equally from traditional pretraining methods. We explore what makes NMT bilingual and BERT LM monolingual representations different on several probing tasks, and why certain training methods succeed in extracting performance from BERT embeddings from NMT tasks.
Biography: 1. Naitian Zhou is a rising junior at the University of Michigan studying computer science and data science. His interests include computational social science and natural language processing.
2. Omar is a Summer 2020 intern with Dr. Jon May. He is also a rising junior at Georgia Institute of Technology.
Host: Jon May and Emily Sheng
More Info: https://nlg.isi.edu/nl-seminar/
Webcast: https://youtu.be/QS3NVSbunt8Location: Information Science Institute (ISI) - Virtual Only
WebCast Link: https://youtu.be/QS3NVSbunt8
Audiences: Everyone Is Invited
Contact: Petet Zamar
Event Link: https://nlg.isi.edu/nl-seminar/
-
NL Seminar-Qualitative Analysis of Unsupervised Neural Machine Translation
Thu, Aug 20, 2020 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Weiqiu You (ISI Intern), Univ. Of Pennsylvania
Talk Title: Qualitative Analysis of Unsupervised Neural Machine Translation
Series: NL Seminar
Abstract: Supervised neural machine translation SNMT models built on all the available parallel data result in higher BLEU on test sets than unsupervised neural machine translation UNMT models leveraging all available monolingual data.
Recently UNMT models such as XLM and MASS are reducing the gap with SNMT in terms of BLEU. Prior work has shown that linguistic and domain dissimilarity often hinder UNMTs performance. The question we're asking in this investigation is, when SNMT and UNMT do have comparable BLEU, do they exhibit qualitative differences, and if these differences can be detected by other metrics.
Biography: Weiqiu You is a rising first year PhD student at University of Pennsylvania who just graduated with MS from UMass Amherst. Her interests include machine translation and natural language processing in general.
Host: Jon May and Emily Sheng
More Info: https://nlg.isi.edu/nl-seminar/
Webcast: https://youtu.be/TawjeAeLbuALocation: Information Science Institute (ISI) - Virtual Only
WebCast Link: https://youtu.be/TawjeAeLbuA
Audiences: Everyone Is Invited
Contact: Petet Zamar
Event Link: https://nlg.isi.edu/nl-seminar/
-
Routing and Task Allocation for Massive Smart-Mobility
Tue, Aug 25, 2020 @ 02:00 PM - 03:00 PM
Conferences, Lectures, & Seminars
Speaker: Dr. Kiril Solovey, Stanford University
Talk Title: Routing and Task Allocation for Massive Smart-Mobility
Abstract: test
Biography: test
Host: Dr. Sven Koenig
Webcast: https://tinyurl.com/y4vuye5rLocation: Online
WebCast Link: https://tinyurl.com/y4vuye5r
Audiences: Everyone Is Invited
Contact: Shao-Hung Chan
-
Routing and Task Allocation for Massive Smart-Mobility
Tue, Aug 25, 2020 @ 02:00 PM - 03:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Dr. Kril Solovey, Stanford University
Talk Title: Routing and Task Allocation for Massive Smart-Mobility
Abstract: Smart mobility holds the promise for transforming our lives with efficient, sustainable, and equitable urban transportation through the incorporation of novel technologies such as the Internet of Things and autonomous vehicles. However, to reap those rewards we must develop control algorithms capable of tackling massive scenarios, while at the same time accounting for their societal impact in terms of congestion, pollution, and fairness, to mention just a few criteria.
In this talk, I will present recent developments in algorithmic design for massive smart-mobility systems in two key areas. First, I will describe an efficient framework for routing fleets of autonomous vehicles while accounting for congestion effects. Then, I will present an approach for multi-drone delivery across broad urban areas, with the end goal of reducing delivery times and mitigating congestion effects of traditional ground-delivery services. A distinguishing feature of the approach is that drones can extend their limited flight range by hitchhiking on public-transit vehicles. In addition to the algorithmic aspects of the two settings, I will discuss the broad implications of such systems to society and highlight directions for future research.
Biography: I am a Postdoctoral Scholar at the Department of Aeronautics and Astronautics, Stanford University, working with Marco Pavone. I obtained a PhD in Computer Science from Tel Aviv University, where I was advised by Dan Halperin.
My research interests lie at the interface of computer science, robotics, and transportation sciences. I pursue deeper insights into the structural properties of fundamental problems in robotics and transportation, for the design of practical algorithmic approaches. My most recent research focuses
on multirobot systems and their application in smart mobility, with an emphasis on societal aspects.
For my work I received multiple awards, including the Clore Scholars and Fulbright Postdoctoral Fellowships, best paper awards and nominations ( 1, 2, 3, 4, 5), and teaching awards. In addition to my academic interests, I take pleasure in being outdoors (mountain biking and hiking), traveling around the world, and enjoy food, cooking, music, books, and meditation.
More information about Kiril's work can be found at kirilsol.github.io
Host: Dr. Sven Koenig
More Info: https://tinyurl.com/y4vuye5r
Webcast: https://tinyurl.com/y4vuye5rLocation: Online
WebCast Link: https://tinyurl.com/y4vuye5r
Audiences: Everyone Is Invited
Contact: Shao-Hung Chan
Event Link: https://tinyurl.com/y4vuye5r
-
ISE 651 - Epstein ISE Seminar
Tue, Aug 25, 2020 @ 03:00 PM - 04:50 PM
Daniel J. Epstein Department of Industrial and Systems Engineering
Conferences, Lectures, & Seminars
Speaker: Prof. Suvrajeet Sen,
Talk Title: Introduction to course by Instructor of Record
Location: Online/Zoom
Audiences: For students only
Contact: Grace Owh
-
AME Seminar
Wed, Aug 26, 2020 @ 03:30 PM - 04:30 PM
Aerospace and Mechanical Engineering
Conferences, Lectures, & Seminars
Speaker: Mahsa Ebrahim, Assistant Professor of Mechanical Engineering Loyola Marymount University
Talk Title: Droplet Impingement and its Application in Electronic Cooling
Abstract: The need for cooling of high heat flux electronics is dramatically increasing as the technology developments make the electronic devices more compact and more powerful. Spray cooling is known to be one of the most effective methods of cooling high heat flux applications. However, common issues of dry-out and excessive liquid accumulation make spray cooling an unreliable technique for the cooling of high heat flux electronics. In order to better understand the fundamental physics and heat absorption of spray cooling, single droplet impingement on a heated surface has gained numerous attention. Although extensive research has been performed on single droplet impingement, the literature is sparse regarding the study of the heat transfer at high impact velocities and at micrometer size scale. Upon the impact of the droplet, the surface temperature drops dramatically and rapidly and at higher impact conditions this temperature drop is even more severe. In this work, the hydrodynamic and heat transfer regimes of a single droplet at high impact conditions were identified and mapped. Techniques were developed to enhance the heat absorption and the evaporation rate of each droplet by utilizing an impinging air flow to increase the impact velocity and to control the hydrodynamic phases after the impact and therefore to promote the utilization of spray cooling in high heat flux electronics.
Biography: Mahsa Ebrahim is an Assistant Professor of Mechanical Engineering at Loyola Marymount University. She received her Ph.D. degree in Mechanical Engineering from Villanova University and her M.S. and B.S. degrees in Mechanical Engineering from K.N.Toosi University of Technology in Iran. She spent a year of her Ph.D. at the University of Leeds in England as a research scholar. Her research focuses on numerical and experimental thermal-fluid science, spray cooling and droplet impingement, interfacial flows and phase interactions with applications in electronic cooling. She has been an active organizer of the IEEE-ITherm conferences since 2018.
Host: AME Department
More Info: https://usc.zoom.us/j/95959519165
Webcast: https://usc.zoom.us/j/95959519165Location: https://usc.zoom.us/j/95959519165
WebCast Link: https://usc.zoom.us/j/95959519165
Audiences: Everyone Is Invited
Contact: Tessa Yao
Event Link: https://usc.zoom.us/j/95959519165
-
ECE Seminar: Understanding Real-World Diffusion Processes: From Epidemics to Violence
Thu, Aug 27, 2020 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Ajitesh Srivastava, Senior Research Associate, Ming Hsieh Department of Computer and Electrical Engineering
Talk Title: Understanding Real-World Diffusion Processes: From Epidemics to Violence
Abstract: The recent outbreak of COVID-19 has encouraged researchers all around the world to identify how their skillsets can be best utilized to contribute to the efforts to deal with the epidemic. Containing the epidemic, providing informed predictions and identifying strategies to restart the economy are essential for the global population to resume their day-to-day life. This may not be the last pandemic we will face, and therefore the research does not end with COVID-19. In fact, the extensive data-collection and monitoring during this epidemic sets the stage for more impactful research in preparedness for future epidemics. Diseases are not the only contagions that fit the above described framework. Violence and drug-abuse are also known to spread through interactions. These issues severely affect the homeless and impede their attempts to safely and successfully exit homelessness and lead a long productive life. California Governor has signed an Executive order this year to assign state resources to reduce homelessness. Modeling how violence and drug-abuse spread among homeless is the key to designing peer-based intervention studies which can assist with these long-term goals.
In this talk, I will discuss utilizing the full potential of Algorithms, Network Science and Data Mining in issues of societal importance, with epidemics and homelessness as examples, and the challenges and opportunities that come with it. By focusing on proper abstractions, we can capture many complexities of the process, and yet come up with a simple model, less prone to overfitting. Doing so can lead to fast and accurate forecasts. For instance, we can train and produce case and death forecasts for more than 3000 counties in under 30 seconds, while still being more accurate than the state-of-the-art methods. Useful modeling informed by sensible assumptions and coupled with optimization approaches can lead to impactful results. For instance, in a Pilot Study at a homeless youth center, we were able to achieve 40% reduction in violence.
Biography: Dr. Ajitesh Srivastava is a Senior Research Associate at Ming Hsieh Department of Computer and Electrical Engineering. He is a member of the Data Science Lab led by Prof. Viktor K. Prasanna. He received his PhD in 2018 from University of Southern California, titled "Computing Cascades: How to Spread Rumors, Win Campaigns, Stop Violence and Predict Epidemics". His research interests include Data Mining, Algorithms, and Network Science applied to epidemics, social good, social networks, architecture, and smart grids. During the COVID-19 pandemic he has been producing weekly forecasts of COVID-19 cases and deaths for the CDC.
Host: Dr. Richard Leahy, leahy@sipi.usc.edu
Webcast: https://usc.zoom.us/j/96802468309?pwd=LzBiWVJCMXI5aUEwM05oTzJXZGQ2QT09WebCast Link: https://usc.zoom.us/j/96802468309?pwd=LzBiWVJCMXI5aUEwM05oTzJXZGQ2QT09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
-
NL Seminar- Translation of Asylum Testimonials from Low-Resource Languages
Thu, Aug 27, 2020 @ 11:00 AM - 12:00 PM
Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Ugur Yavuz ISI Summer Intern, Dartmouth College
Talk Title: Translation of Asylum Testimonials from Low-Resource Languages
Abstract: Many asylum seekers along the southern border of the United States speak low-resource languages that are not available on commercial translation services. As a result, the translation of their testimonials poses a real challenge to the legal system, as well as non-governmental organizations. We will discuss potential techniques that would facilitate this task, such as transfer learning, domain adaptation, and corpus expansion, and also explain our work in compiling bilingual corpora for Mixtec and Kanjobal languages.
Biography: Ugur Yavuz is a summer intern working at the Natural Language Group with Dr. Jon May. He is a rising senior at Dartmouth College studying computer science and mathematics.
Host: Jon May and Emily Sheng
More Info: https://nlg.isi.edu/nl-seminar/
Webcast: https://youtu.be/xL9kMb2lTa4Location: Information Science Institute (ISI) - Virtual Only
WebCast Link: https://youtu.be/xL9kMb2lTa4
Audiences: Everyone Is Invited
Contact: Petet Zamar
Event Link: https://nlg.isi.edu/nl-seminar/
