Select a calendar:
Filter December Events by Event Type:
SUNMONTUEWEDTHUFRISAT
Events for December 08, 2021
-
USC AI Futures Symposium on AI with Common SenseWed, Dec 08, 2021 @ 09:00 AM - 12:00 PM
Thomas Lord Department of Computer Science, Information Sciences Institute
Conferences, Lectures, & Seminars
Speaker: Multiple panelists, USC and collaborators
Talk Title: AI with Common Sense
Series: USC AI Futures Symposium on AI
Abstract: Humans use common sense knowledge to understand how the world works. AI systems still lack common sense knowledge, which is critical in many domains in order to form expectations, manage unexpected situations, and connect with the human experience. This symposium presents an overview of research at USC on acquiring and organizing commonsense knowledge, integrating it into AI systems, and measuring its impact on improving AI system behaviors and interactions with people.
This event is part of the USC AI Futures Symposium Series. Prior events were held in May 2021 with the theme: AI and Data Science, and in January 2021 with the theme: Will AIs Ever Be One of Us?.
Please visit website to register and view updated information https://isi-usc-edu.github.io/USC-CommonSense-Symposium/
Host: Yolanda Gil
Location: Virtual
Audiences: Everyone Is Invited
Contact: Lori Weiss/USC ISI
-
MS CSCI/DSCI Drop-in Academic Advising
Wed, Dec 08, 2021 @ 10:00 AM - 11:00 AM
Thomas Lord Department of Computer Science
Workshops & Infosessions
Virtual Academic Advising Drop-in Hours for continuing MS students in CS or Data Science Programs will be available for the month of December. If you have a quick question that doesn't require a 20 minute appointment, please come to our drop in hours. Students may be placed into the waiting room upon arrival.
Zoom access link for all sessions:
ZOOM LINK SENT TO STUDENTS DIRECTLY. CHECK EMAIL FOR LINK.
Wednesday, December 1st --- 10am -- 11am
Wednesday, December 1st --- 2:30pm -- 3:30pm
Wednesday, December 8th --- 10am -- 11am
Wednesday, December 8th --- 2:30pm -- 3:30pm
Wednesday, December 15th --- 10am -- 11am
Wednesday, December 15th --- 2:30pm -- 3:30pmLocation: Online
Audiences: Graduate
Contact: USC Computer Science
-
Computer Science General Faculty Meeting
Wed, Dec 08, 2021 @ 12:00 PM - 02:00 PM
Thomas Lord Department of Computer Science
Receptions & Special Events
Bi-Weekly regular faculty meeting for invited full-time Computer Science faculty only. Event details emailed directly to attendees.
Location: TBD
Audiences: Invited Faculty Only
Contact: Assistant to CS chair
-
PhD Thesis Proposal - Hikaru Ibayashi
Wed, Dec 08, 2021 @ 02:00 PM - 03:30 PM
Thomas Lord Department of Computer Science
University Calendar
PhD Thesis Proposal - Hikaru Ibayashi
Wed, Dec 8, 2021 @ 02:00 PM - 03:30 PM
Committee members: Chair: Prof. Aiichiro Nakano, Prof. Satish Kumar Thittamaranahalli, Prof. Emilio Ferrara, Prof. Yan Liu, Prof. Paulo Branicio (Department of Chemical Engineering and Materials Science)
Title:
Sharpness analysis of neural-networks for high-performance physics simulation
Abstract:
In recent years, deep neural networks have witnessed tremendous success in a wide range of fields. Especially, in physics simulations, neural networks have achieved drastically efficient computation by approximating first-principles simulations. However, such practical successes have opened theoretical problems. One open question is why a simple optimization algorithm such as stochastic gradient descent (SGD) can find solutions that generalize well over non-convex loss surfaces. In this proposal, we leverage the second-order information of loss surface, i.e., sharpness, to lay a theoretical foundation of the generalizability of neural networks. First, we use a novel quasi-potential theory to prove that SGD avoids non-generalizing sharp minima. Secondly, we develop a scale-invariant sharpness measure named "minimum sharpness" to theoretically explain why sharp minima are not generalizing. Finally, as a practical application of the thus-developed theoretical framework, we propose a novel sharpness-regularization scheme for robust neural-network-based molecular dynamics simulations. This research will demonstrate the effectiveness of sharpness analysis to deepen the understanding of neural networks and their successful application in physics simulations.
Zoom link: https://usc.zoom.us/j/7751892842WebCast Link: https://usc.zoom.us/j/7751892842
Audiences: Everyone Is Invited
Contact: Lizsl De Leon
-
MS CSCI/DSCI Drop-in Academic Advising
Wed, Dec 08, 2021 @ 02:30 PM - 03:30 PM
Thomas Lord Department of Computer Science
Workshops & Infosessions
Virtual Academic Advising Drop-in Hours for continuing MS students in CS or Data Science Programs will be available for the month of December. If you have a quick question that doesn't require a 20 minute appointment, please come to our drop in hours. Students may be placed into the waiting room upon arrival.
Zoom access link for all sessions:
ZOOM LINK SENT TO STUDENTS DIRECTLY. CHECK EMAIL FOR LINK.
Wednesday, December 1st --- 10am -- 11am
Wednesday, December 1st --- 2:30pm -- 3:30pm
Wednesday, December 8th --- 10am -- 11am
Wednesday, December 8th --- 2:30pm -- 3:30pm
Wednesday, December 15th --- 10am -- 11am
Wednesday, December 15th --- 2:30pm -- 3:30pm
Location: Online
Audiences: Graduate
Contact: USC Computer Science
-
CS Colloquium: Yajie Zhao (USC ICT) - Automating 3D Contents Creation for AR/VR
Wed, Dec 08, 2021 @ 03:30 PM - 04:30 PM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Yajie Zhao, USC
Talk Title: Automating 3D Contents Creation for AR/VR
Abstract: The last couple of years have shown an incredible boost in performance, quality of real-time applications of AR/VR. The creation of high quality 3D assets, including virtual human, objects and 3D environments is tied with complex capture equipment, massive data, a long production cycle and intensive manual labor by a professional team. And it may still be in the notorious Uncanny Valley. In this talk, I will explore how to produce high quality 3D assets in a low-cost way. I will discuss how to leverage AI-based technologies to automate, accelerate, and simplify the industrial production procedure of 3D contents creation from data capturing to rendering. And bring photorealism to the next level!
This lecture satisfies requirements for CSCI 591: Research Colloquium
Join Zoom Meeting
https://usc.zoom.us/j/97098103729
Meeting ID: 970 9810 3729
Biography: Dr. Yajie Zhao is a computer scientist at USC Institute for Creative Technologies (ICT), University of Southern California. She is currently the acting director of the Vision and Graphics Lab of USC-ICT. Her research interests are high-quality 3D content creation for AR/VR, which includes human digitization, performance capturing, and scene reconstruction/ understanding. Yajie Zhao earned her Ph.D. degree in 2017 from the University of Kentucky under the supervision of Dr. Ruigang Yang.
Host: Aram Galstyan
Audiences: Everyone Is Invited
Contact: Cherie Carter
