Select a calendar:
Filter February Events by Event Type:
Events for February 03, 2025
-
EiS Communications Hub - Tutoring for Engineering Ph.D. Students
Mon, Feb 03, 2025 @ 10:00 AM - 12:00 PM
Viterbi School of Engineering Student Affairs
Workshops & Infosessions
Viterbi Ph.D. students are invited to drop by the Hub for instruction on their writing and speaking tasks! All tutoring is one-on-one and conducted by Viterbi faculty.
Location: Ronald Tutor Hall of Engineering (RTH) - 222A
Audiences: Viterbi Ph.D. Students
Contact: Helen Choi
Event Link: https://sites.google.com/usc.edu/eishub/home
-
CS Colloquium: Justin Solomon (MIT) - Navigating, Restructuring and Reshaping Learned Latent Spaces
Mon, Feb 03, 2025 @ 10:00 AM - 11:00 AM
Thomas Lord Department of Computer Science
Conferences, Lectures, & Seminars
Speaker: Justin Solomon, MIT
Talk Title: Navigating, Restructuring and Reshaping Learned Latent Spaces
Abstract: Modern machine learning architectures often embed their inputs into a lower-dimensional latent space before generating a final output. A vast set of empirical results---and some emerging theory---predicts that these lower-dimensional codes often are highly structured, capturing lower-dimensional variation in the data. Based on this observation, in this talk I will describe efforts in my group to develop lightweight algorithms that navigate, restructure, and reshape learned latent spaces. Along the way, I will consider a variety of practical problems in machine learning, including low-rank adaptation of large models, regularization to promote local latent structure, and efficient training/evaluation of generative models. This talk will cover collaborative research with Rickard Gabrielsson, Kimia Nadjahi, Chris Scarvelis, Tal Shnitzer, Mikhail Yurochkin, Jiacheng Zhu, and others.
This lecture satisfies requirements for CSCI 591: Research Colloquium
Biography: Justin Solomon is an Associate Professor of Electrical Engineering and Computer Science at MIT. He leads the Geometric Data Processing Group in the MIT Computer Science and Artificial Intelligence Laboratory (CSAIL), which studies problems at the intersection of geometry, large-scale optimization, and applications.
Host: Yue Wang
Location: Olin Hall of Engineering (OHE) - 132
Audiences: Everyone (USC) is invited
Contact: CS Faculty Affairs
-
PhD Thesis Proposal - Tingting Tang
Mon, Feb 03, 2025 @ 12:30 PM - 01:30 PM
Thomas Lord Department of Computer Science
University Calendar
Title: Optimizing Privacy-Preserving Machine Learning for Improved Privacy, Utility, and Efficiency Tradeoffs
Location: EEB 349
Date and Time: February 3, 2025, 12.30 PM-1.30 PM
Zoom Link: https://usc.zoom.us/j/7995244109?pwd=OUp6RWhUZlFGclgyN3hkREh0Z21ldz09
Committee: Murali Annavaram (Chair), Salman Avestimehr, Bhaskar Krishnamachari, Harsha Madhyastha, Sai Praneeth Karimireddy
Abstract: Privacy-preserving machine learning (PPML) is essential for protecting sensitive data in machine-learning applications, requiring a careful balance between privacy, utility, and efficiency. However, the trade-offs and interdependencies among these dimensions present significant design challenges. This thesis proposal explores and optimizes their interplay through low-rank decomposition, focusing on two key PPML technologies: Differential Privacy (DP) and Secure Multiparty Computation (MPC). In the context of DP-based graph neural networks (GNNs), I propose a novel training framework leveraging low-rank singular value perturbation to protect sensitive graph edges while preserving the primary graph structure. This approach achieves a significantly improved privacy-utility trade-off and demonstrates resilience to edge inference attacks. For MPC-based secure model inference, I propose leveraging low-rank decomposition for the linear layers of ML models, reducing the number of MPC multiplications required during offline and online phases. Techniques such as truncation skipping and linear layer concatenation further reduce computational and communication overheads, enhancing overall efficiency in MPC ML workflows without compromising the robust security guarantees provided by MPC. By addressing the interactions between privacy, utility, and efficiency, my proposal lays the foundation for more practical and effective deployment of privacy-preserving machine learning solutions in real-world applications.Location: Hughes Aircraft Electrical Engineering Center (EEB) - 349
Audiences: Everyone Is Invited
Contact: Tingting Tang
-
Photonics Seminar - Alexander Szameit, Monday, February 3rd at 2pm in EEB 248
Mon, Feb 03, 2025 @ 02:00 PM - 03:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Alexander Szameit, Professor, Chair for Experimental Solid-State Optics, University of Rostock
Talk Title: Topology in space, time, and space-time
Series: Photonics Seminar Series
Abstract: In recent years, topological phenomena in photonic systems have attracted much attention, with their striking features arising from robust states in the energy gaps of spatially periodic media. However, light waves are entities that extend in space as well as time, such that one may ask whether topological effects can also occur in the temporal domain, or even space-time. Intuitively, systems that are periodic in time may be gapped in momentum, leading to topological states localized at time interfaces. However, time - in contrast to space - exhibits a unique unidirectionality often referred to as the "arrow of time". Inspired by these features, I will present our most recent experiments on topological states residing at temporal interfaces. Moreover, I will discuss the formation of spacetime-topological events and demonstrate unique features such as their limited collapse under disorder and causality-suppressed coupling.
Biography: Alexander Szameit (*1979 in Halle, Germany) studied Physics at the Universities of Halle and Jena, Germany. He obtained his Diploma and PhD in 2004 and 2007, respectively. After spending time in Australia and Israel, he returned to Jena as an Assistant Professor in 2011. After receiving his habilitation in 2015, he was appointed as Full Professor at the University of Rostock in 2016, where he holds the chair for Experimental Solid-State Optics. His work deals with all aspects of complex light evolution in large-scale integrated photonic waveguide circuits, with a particular focus on topological photonics.
Host: Mercedeh Khajavikhan and Demetri Christodoulides
More Information: Alexander Szameit Flyer.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Marilyn Poplawski
-
Computational Science Distinguished Seminar
Mon, Feb 03, 2025 @ 04:00 PM - 05:00 PM
USC School of Advanced Computing
Conferences, Lectures, & Seminars
Speaker: George Haller, ETH Zürich
Talk Title: Nonlinear Spectral Model Reduction from Data
Abstract: Machine learning has been a major development in applied science and engineering, with impressive success stories in static learning environments like image, pattern, and speech recognition. Yet the modeling of dynamical phenomena—such as nonlinear vibrations of solids and transitions in fluids—remains a challenge for classic machine learning. Indeed, neural net models for nonlinear dynamics tend to be complex, uninterpretable and unreliable outside their training range.
In this talk, I discuss a dynamical systems alternative to neural networks in the data-driven reduced-order modeling of nonlinear phenomena. Specifically, I show that the recent concept of spectral submanifolds (SSMs) provides very low-dimensional attractors in a large family of mechanics problems ranging from wing oscillations to transitions in shear flows. A data-driven identification of the reduced dynamics on these SSMs gives a mathematically justified way to construct accurate and predictive reduced-order models for solids, fluids and controls without the use of governing equations. I illustrate this on physical problems including the accelerated finite-element simulations of large structures, prediction of transitions to turbulence, reduced-order modeling of fluid-structure interactions, extraction of reduced equations of motion from videos, and model-predictive control of soft robots.
Biography: George Haller is a professor of Mechanical Engineering at ETH Zürich, where he holds the Chair in Nonlinear Dynamics and heads the Institute for Mechanical Systems. His prior appointments include tenured faculty positions at Brown, McGill and MIT. He also served as the inaugural director of Morgan Stanley’s fixed income modeling center. Professor Haller is a recipient of a Sloan Fellowship in mathematics, an ASME Thomas Hughes Young Investigator Award, a School of Engineering Distinguished Professorhip (McGill), and the Stanley Corrsin Award of the APS. He is an external member of the Hungarian Academy of Science and an elected fellow of SIAM, APS and ASME. He currently serves as feature editor at Nonlinear Dynamics and senior editor at the Journal of Nonlinear Science. His research focuses on nonlinear dynamical systems with applications to mechanical vibrations, coherent structures in turbulence, and data- and equation-driven model reduction for physical systems. He has authored three monographs in these areas.
Host: School of Advanced Computing
More Info: https://sac.usc.edu/events/
Location: Ronald Tutor Hall of Engineering (RTH) - 526
Audiences: Everyone Is Invited
Contact: Tessa Yao
Event Link: https://sac.usc.edu/events/