Select a calendar:
Filter November Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Conferences, Lectures, & Seminars
Events for November
-
CS Colloquium: Yuanzhi Li (CMU) - Multi-player Multi-armed Bandit: Can We Collaborate Without "Zoom"?
Tue, Nov 03, 2020 @ 03:30 PM - 04:30 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Yuanzhi Li, Carnegie Mellon University
Talk Title: Multi-player Multi-armed Bandit: Can We Collaborate Without "Zoom"?
Series: Computer Science Colloquium
Abstract: Multi-armed bandit is a well-established area in online decision making, where one player makes sequential decisions in a non-stationary environment to maximize his/her accumulative rewards. The traditional multi-armed bandit problem becomes significantly more challenging when there are multiple players in the same environment, while only one piece of reward is presented at a time for each arm. In this setting, if two players pick the same arm at the same round, they are only able to get one piece of reward instead of two. When the rewards are non-negative, to maximize the total accumulative rewards by all players, they need to collaborate to avoid "collision" -- i.e. the players need to make sure that they do not all rush to the same arm (even if it has the highest reward) at the same round. We focus on the setting where communications between players are completely disabled: e.g. they are separated in different places of the world without any "Zoom". We show that low-regret can still be obtained in this setting: Players can actually collaborate to maximize total rewards by avoiding collision in a non-stationary environment, even when they do not communicate at all during the entire sequence of decisions.
Register in advance for this webinar at:
https://usc.zoom.us/webinar/register/WN_kVp5jz5qSIKAZIphNGWaWw
After registering, attendees will receive a confirmation email containing information about joining the webinar.
This lecture satisfies requirements for CSCI 591: Research Colloquium.
Biography: Yuanzhi Li is an assistant professor at CMU, Machine Learning Department. He did his Ph.D. at Princeton, under the advice of Sanjeev Arora (2014-2018) as well as a one-year postdoc at Stanford. His wife is Yandi Jin.
Host: Haipeng Luo
More Info: https://usc.zoom.us/webinar/register/WN_kVp5jz5qSIKAZIphNGWaWw
Location: Online Zoom Webinar
Audiences: Everyone Is Invited
Contact: Computer Science Department
Event Link: https://usc.zoom.us/webinar/register/WN_kVp5jz5qSIKAZIphNGWaWw
-
WiE's Negotiation Seminar with Tahl Raz, 11/4 at 1pm
Wed, Nov 04, 2020 @ 01:00 PM - 02:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Tahl Raz, Co-Author of Never Split the Difference
Talk Title: Negotiation Seminar
Abstract: Our event is back on! We're happy that our speaker is feeling better and are looking forward to seeing you all soon! The Graduate Committee of Women in Engineering is excited to host Tahl Raz, New York Times bestselling author, award-winning journalist, and co-author of the nation's leading publication on negotiation, Never Split the Difference, in our Negotiation Seminar on Wednesday, November 4th at 1pm PST.
RSVP to attend!
Learn more about Tahl: https://www.linkedin.com/in/tahlraz
Zoom Link: https://usc.zoom.us/j/98308499819?pwd=bVZHeDJRODcrSlFpN3hGZ1dyczU2UT09
RSVP Form: https://forms.gle/7dHxaaMwyq7faceb8
Biography: Learn more about Tahl: https://www.linkedin.com/in/tahlraz
Host: The Graduate Committee of Women in Engineering
More Info: https://forms.gle/7dHxaaMwyq7faceb8
Webcast: https://usc.zoom.us/j/98308499819?pwd=bVZHeDJRODcrSlFpN3hGZ1dyczU2UT09Location: Zoom
WebCast Link: https://usc.zoom.us/j/98308499819?pwd=bVZHeDJRODcrSlFpN3hGZ1dyczU2UT09
Audiences: Everyone Is Invited
Contact: USC Computer Science
Event Link: https://forms.gle/7dHxaaMwyq7faceb8
-
CS Colloquium: Xuezhe Ma (USC ISI) - Towards Structured-Infused and Disentangled Representation Learning
Tue, Nov 10, 2020 @ 11:00 AM - 12:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Xuezhe Ma, USC
Talk Title: Towards Structured-Infused and Disentangled Representation Learning
Abstract: One of the keys to the empirical successes of deep neural networks in many domains, such as natural language processing and computer vision, is their ability to automatically extract salient features for downstream tasks via the end-to-end learning paradigm.
In this talk, I will present two of our recent work. First, I will introduce how to encode structured dependencies into learned representations to achieve efficient non-autoregressive machine translation models. Second, I will present our work on learning representations to decouple global and local information from/for image generation. I will conclude by laying out future research directions towards interpretable and controllable representation learning.
This lecture satisfies requirements for CSCI 591: Research Colloquium
Join Zoom Meeting
https://usc.zoom.us/j/91743613540?pwd=S0hPWEk5MHFSTVdoSmVidkxLVmlwQT09
Meeting ID: 917 4361 3540
Passcode: 296867
Biography: Xuezhe Ma joined ISI as a computer scientist in Fall 2020.
Xuezhe received his PhD degree in Language Technologies Institute at Carnegie Mellon University, advised by Eduard Hovy.
Before that, he received his B.E and M.S from Shanghai Jiao Tong University. His research interests fall in areas of natural language processing and machine learning, particularly in deep learning and representation learning with applications to linguistic structured prediction and deep generative models. Xuezhe has interned at Allen Institute for Artificial Intelligence (AI2) and earned the AI2 Outstanding Intern award. His research has been recognized with outstanding paper award at ACL 2016 and best demo paper nomination at ACL 2019.
Host: Xiang Ren
Audiences: Everyone Is Invited
Contact: Cherie Carter
-
CS Distinguished Lecture: Steve Easterbrook (University of Toronto) - Computing the Climate: Building the Software for Understanding Climate Change
Tue, Nov 10, 2020 @ 03:30 PM - 04:50 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Steve Easterbrook, University of Toronto
Talk Title: Computing the Climate: Building the Software for Understanding Climate Change
Series: Computer Science Distinguished Lecture Series
Abstract: The history of climate science is closely tied to the history of computing. Climate scientists have always pushed the limits of computational modelling, from the first computational weather forecasts developed by von Neumann and Charney to run on ENIAC, to the earth system models used to produce projections of future climate change for the most recent IPCC reports. Along the way, climate scientists have developed a sophisticated set of software development practices tailored to the needs of a science in which virtual experiments are essential for understanding the relationships between human activity and the global climate system. In this talk, I will first explain what climate models do, via a quick tour of the history of climate modelling. I will then show how a core set of software development practices are used to support a culture of scientific experimentation which provides robust answers to societally important questions. I will end the talk with a brief overview of the current generation of climate model experiments. These address critically important questions such as whether there are still viable pathways to deliver the UN's commitment to constrain global warming to no more than +2*C, and whether geo-engineering can buy us more time to address the underlying causes of climate change.
Register in advance for this webinar at:
https://usc.zoom.us/webinar/register/WN_0sw0PJhSTFuyqKxoQie5Gw
After registering, attendees will receive a confirmation email containing information about joining the webinar.
This lecture satisfies requirements for CSCI 591: Research Colloquium.
Biography: Steve Easterbrook is the Director of the School of the Environment and Professor of Computer Science at the University of Toronto. He received his Ph.D. (1991) in Computing from Imperial College in London (UK), and joined the faculty at the School of Cognitive and Computing Science, University of Sussex. From 1995-99, he was lead scientist at NASA's Independent Verification and Validation (IV&V) Facility in West Virginia, where he investigated software verification on the Space Shuttle Flight Software, the International Space Station, and the Earth Observation System. He moved to the University of Toronto in 1999. His research interests range from modelling and analysis of complex adaptive systems to the socio-cognitive aspects of team interaction. His current research is in climate informatics, where he studies how climate scientists develop computational models to improve their understanding of earth systems and climate change, and the broader question of how that knowledge is shared with other communities. He has been a visiting scientist at the UK Met Office Hadley Centre, in Exeter, the National Centre for Atmospheric Research in Boulder, Colorado; the Max-Planck Institute for Meteorology, in Hamburg, and the Institute Pierre Simon Laplace in Paris.
Host: Heather Culbertson
More Info: https://usc.zoom.us/webinar/register/WN_0sw0PJhSTFuyqKxoQie5Gw
Location: Online Zoom Webinar
Audiences: Everyone Is Invited
Contact: Computer Science Department
Event Link: https://usc.zoom.us/webinar/register/WN_0sw0PJhSTFuyqKxoQie5Gw
-
CS Colloquium: Muhao Chen (USC ISI) - Knowledge Acquisition with Transferable Representation Learning
Thu, Nov 12, 2020 @ 03:30 PM - 04:30 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Muhao Chen, USC
Talk Title: Knowledge Acquisition with Transferable Representation Learning
Abstract: Multi-relational data provide structural and actionable knowledge representations for various AI systems. As constructing such structural knowledge is often costly and has relied on extensive human effort, there is a pressing need for approaches to automate knowledge acquisition. In this talk, I will summarize two lines of my research to accomplish this mission: (i) transferable representation learning, and (ii) constrained and indirect supervision. Transferable representation learning can automatically capture the association of knowledge across different data sources with minimal supervision, therefore holds the promise of creating a universal representation scheme to support the synchronization of knowledge. Meanwhile, constrained and indirect supervision methods could develop more reliable learning systems for knowledge acquisition from unstructured data, particularly in cases without sufficient training labels. Based on these two lines of research, I will also discuss several applications for a wide range of tasks in areas of knowledge base construction, natural language understanding and computational biology.
This talk satisfies requirements for CSCI 591: Research Colloquium
Join Zoom Meeting
https://usc.zoom.us/j/96706950791?pwd=cXp3TWlhRmo5ZDB0bnA0a0lOQ1VVdz09
Meeting ID: 967 0695 0791
Passcode: 808248
Biography: Muhao Chen joined as a computer scientist at USC ISI in Fall 2020. Prior to that, he was a postdoctoral fellow at UPenn, hosted by Dan Roth. He received his Ph.D. in Computer Science from UCLA in 2019, and B.S. in Computer Science from Fudan University in 2014. His research focuses on data-driven machine learning approaches for processing structured data, and knowledge acquisition from unstructured data. Particularly, he is interested in developing knowledge-aware learning systems with generalizability and requiring minimal supervision, and with concrete applications to natural language understanding, knowledge base construction, computational biology and medicine. Muhao has published over 40 papers in leading AI, NLP and Comp. Bio/med venues. His work has received a best student paper award at ACM BCB, and best paper award nomination at CoNLL. Additional information is available at https://muhaochen.github.io/
Host: CS Department
Audiences: Everyone Is Invited
Contact: Cherie Carter
-
CS Colloquium: Mohammad Rostami (USC ISI) - Learning Efficiently in Data-Scarce Regimes
Tue, Nov 17, 2020 @ 11:00 AM - 12:00 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Mohammad Rostami, USC
Talk Title: Learning Efficiently in Data-Scarce Regimes
Abstract: The unprecedented processing demand, posed by the explosion of big data, challenges researchers to design efficient and adaptive machine learning algorithms that do not require persistent retraining and avoid learning redundant information. Inspired from learning techniques of intelligent biological agents, identifying transferable knowledge across learning problems has been a significant research focus to improve machine learning algorithms. In this talk, we explain how the challenges of knowledge transfer can be addressed through embedding spaces that capture and store hierarchical knowledge.
We first focus on the problem of cross-domain knowledge transfer. We explore the problem of zero-shot image classification, where the goal is to identify images from unseen classes using semantic descriptions of these classes. We train two coupled dictionaries that align visual and semantic domains via an intermediate embedding space. We then extend this idea by training deep networks that match data distributions of two visual domains in a shared cross-domain embedding space.
We then investigate the problem of cross-task knowledge transfer in sequential learning settings. Here, the goal is to identify relations and similarities of multiple machine learning tasks to improve performance across the tasks. We first address the problem of zero-shot learning in a lifelong machine learning setting, where the goal is to learn tasks with no data using high-level task descriptions. Our idea is to relate high-level task descriptors to the optimal task parameters through an embedding space. We then develop a method to overcome the problem of catastrophic forgetting within a continual learning setting of deep neural networks by enforcing the tasks to share the same distribution in the embedding space.
Finally, we focus on current research directions to expand the past progress and plans for the future research directions. Through this talk, we demonstrate that despite major differences, problems within the above learning scenarios can be tackled using a unifying strategy that allows transferring knowledge effectively.
This lecture satisfies requirements for CSCI 591: Research Colloquium
Join Zoom Meeting
https://usc.zoom.us/j/91954313931?pwd=U3JmUWR4WVZ6aDEyMUs0dEk0akZ5QT09
Meeting ID: 919 5431 3931
Passcode: 299776
Biography: Mohammad Rostami is a computer scientist at USC Information Sciences Institute. He received Ph.D. degree in Electrical and Systems Engineering from the University of Pennsylvania in August 2019. He also received an M.S. degree in Robotics and M.A. degree in Philosophy at Penn. Before Penn, he obtained an M.Sc. degree in Electrical and Computer Engineering from University of Waterloo, and B.Sc. degree in Electrical Engineering and B.Sc. degree in Mathematics from the Sharif University of Technology. His current research area is learning in time-dependent and data-scarce regimes within machine learning.
Host: CS Department
Audiences: Everyone Is Invited
Contact: Cherie Carter
-
CS Distinguished Lecture: Jennifer Rexford (Princeton University) - Securing Internet Applications From Routing Attacks
Tue, Nov 17, 2020 @ 03:30 PM - 04:50 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Jennifer Rexford, Princeton University
Talk Title: Securing Internet Applications From Routing Attacks
Series: Computer Science Distinguished Lecture Series
Abstract: The Internet is a "network of networks" that interconnects tens of thousands of separately administered networks. Yet, the Border Gateway Protocol (BGP), the glue that holds the disparate parts of the Internet together, is notoriously vulnerable to misconfiguration and attack. The consequences range from making destinations unreachable, to misdirecting traffic through unexpected intermediaries, to impersonating legitimate services. Attacks on Internet routing are typically viewed through the lens of availability and confidentiality, assuming an adversary that either discards traffic or performs eavesdropping. Yet, a strategic adversary can use routing attacks to compromise the security of critical Internet applications like Tor, certificate authorities, and the bitcoin network. In this talk, we survey such application-specific routing attacks and argue that both application-layer and network-layer defenses are essential and urgently needed. While application-layer defenses are easier to deploy in the short term, we hope that greater awareness of strategic attacks on important applications can provide much needed momentum for the deployment of network-layer defenses like secure routing protocols.
Register in advance for this webinar at:
https://usc.zoom.us/webinar/register/WN_uiLYEP8mRR2_UIQ4oJn5ug
After registering, attendees will receive a confirmation email containing information about joining the webinar.
This lecture satisfies requirements for CSCI 591: Research Colloquium.
Biography: Jennifer Rexford is the Gordon Y.S. Wu Professor of Engineering and the Chair of Computer Science at Princeton University. Before joining Princeton in 2005, she worked for nine years at AT&T Labs--Research. Jennifer received her BSE degree in electrical engineering from Princeton University in 1991, and her PhD degree in electrical engineering and computer science from the University of Michigan in 1996. She is co-author of the book "Web Protocols and Practice" (Addison-Wesley, 2001). She served as the chair of ACM SIGCOMM from 2003 to 2007. Jennifer received ACM's Grace Murray Hopper Award for outstanding young computer professional, the ACM Athena Lecturer Award, the NCWIT Harrold and Notkin Research and Graduate Mentoring Award, the ACM SIGCOMM award for lifetime contributions, and the IEEE Internet Award. She is an ACM Fellow, an IEEE Fellow, and a member of the American Academy of Arts and Sciences, the National Academy of Engineering, and the National Academy of Sciences.
Host: Heather Culbertson
More Info: https://usc.zoom.us/webinar/register/WN_uiLYEP8mRR2_UIQ4oJn5ug
Location: Online Zoom Webinar
Audiences: Everyone Is Invited
Contact: Computer Science Department
Event Link: https://usc.zoom.us/webinar/register/WN_uiLYEP8mRR2_UIQ4oJn5ug
-
CS Colloquium: Matthew Gombolay (Georgia Institute of Technology) - Democratizing Robot Learning for Safe, Efficient Human-Robot Interaction
Thu, Nov 19, 2020 @ 03:30 PM - 04:50 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Matthew Gombolay, Georgia Institute of Technology
Talk Title: Democratizing Robot Learning for Safe, Efficient Human-Robot Interaction
Series: Computer Science Colloquium
Abstract: Robotic technology offers the promise of performing at-home care tasks, revitalizing manufacturing, and even scaling the power of earth-bound scientists in autonomous space exploration. However, each new robot deployment today requires an ad hoc army of consultants and vast computing resources operating on black box, sample-inefficient models. To unlock the potential of robotics, we need to democratize machine learning and put the power of these tools in the hands of the end user. In this talk, I will present exciting, novel work in my lab that enables to safely and efficiently learn from human teachers and interactions with their environments. I will demonstrate how we can 1) enable robots to learn new skills from heterogeneous human teachers, 2) balance the need to actively learn more about their environment while remaining safe in proximity to humans, and 3) and convey their knowledge to human teachers and teammates through interpretable machine learning representations.
Register in advance for this webinar at:
https://usc.zoom.us/webinar/register/WN_6Ti2CLNuS7SqIcROZ7FJ6Q
After registering, attendees will receive a confirmation email containing information about joining the webinar.
This lecture satisfies requirements for CSCI 591: Research Colloquium.
Biography: Dr. Matthew Gombolay is an Assistant Professor of Interactive Computing at the Georgia Institute of Technology. He received a B.S. in Mechanical Engineering from the Johns Hopkins University in 2011, an S.M. in Aeronautics and Astronautics from MIT in 2013, and a Ph.D. in Autonomous Systems from MIT in 2017. Gombolay's research interests span robotics, AI/ML, human-robot interaction, and operations research. Between defending his dissertation and joining the faculty at Georgia Tech, Dr. Gombolay served as a technical staff member at MIT Lincoln Laboratory, transitioning his research to the U.S. Navy, earning him an R&D 100 Award. His publication record includes a best paper award from American Institute for Aeronautics and Astronautics, a best student paper from the American Controls Conference, and he was selected as a DARPA Riser in 2018. He was also awarded a NASA Early Career Fellowship for his work increasing science autonomy in space.
Host: Stefanos Nikolaidis
More Info: https://usc.zoom.us/webinar/register/WN_6Ti2CLNuS7SqIcROZ7FJ6Q
Location: Online Zoom Webinar
Audiences: Everyone Is Invited
Contact: Computer Science Department
Event Link: https://usc.zoom.us/webinar/register/WN_6Ti2CLNuS7SqIcROZ7FJ6Q
