SUNMONTUEWEDTHUFRISAT
Events for February
-
Photonics SEminar - Jie Qiao, Friday, Feb 2nd at 3pm in EEB 248
Fri, Feb 02, 2024 @ 03:00 PM - 04:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Jie Qiao, Rochester Institute of Technology
Talk Title: Ultrafast-Lasers-Enabled Photonics, Optics, and Waveguide Lasers
Series: Photonics Seminar Series
Abstract: The investigation into ultrafast-laser-based photonics fabrication and integration represents multifaceted interdisciplinary research, intersecting applied physics, photonics, lasers, materials, and imaging. This presentation describes computational models and elucidates physical processes pertaining to the utilization of ultrafast lasers for the fabrication of optical, photonic, and laser components. Topics covered include the 3D writing of waveguides, waveguide lasers, and beam splitters in crystal and glass materials, as well as nanostructuring, shape correction, and the precision bonding of semiconductor and dielectric materials.
Biography: Dr. Jie Qiao is an associate professor at the Carlson Center for Imaging Science at the Rochester Institute of Technology. Her research at RIT focuses on ultrafast laser phonics, wavefront sensing and beam shaping. Prior to joining RIT, she was a laser system scientist at the Department -of-Energy-funded Laboratory for Laser Energetics, the University of Rochester. She led the demonstration of the world's first 1.5-meter coherently-phased-grating pulse compressor for the OMEGA EP kilojoule, petawatt lasers. She has worked on technology innovation of various ultrafast laser systems, photonics devices, optical imaging, and metrology systems for two photonic startups and one optics company. She was a Fulbright US research scholar and a visiting professor at the Center for Intense Lasers and Applications (CELIA), Universite Bordeaux, France in the 2022 academic year. Dr. Qiao is an Optica Fellow and was an associate editor for Optics Express from 2018 to 2021. She is the General Chair for the 2024 and 2025 CLEO conference, the Application and Technology Program. She earned her doctoral degree from the Department of Electrical and Computer Engineering, University of Texas, Austin.
Host: Mercedeh Khajavikhan, Michelle Povinelli, Constantine Sideris; Hossein Hashemi; Wade Hsu; Mengjie Yu; Wei Wu; Tony Levi; Alan E. Willner; Andrea Martin Armani
More Information: Jie Qiao Seminar.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Marilyn Poplawski
-
Semiconductors & Microelectronics Technology Seminar - Tingyi Gu, Friday, Feb. 9th at 2pm in EEB 248
Fri, Feb 09, 2024 @ 02:00 PM - 03:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Tingyi Gu, ECE- University of Delaware
Talk Title: On-chip wavefront shaping and image classification on silicon photonics
Series: Semiconductors & Microelectronics Technology
Abstract: The advancement of nanotechnologies enables powerful control of photons by subwavelength structures. In recent years, rapid advancement of metasurface and metamaterials reveal the potential of nanophotonics in the applications across disciplines, from image processing/conversion to controlled light-matter interactions. In this talk, I will progressively illustrate the powerful role of the meta-atoms, meta-surface, and meta-system in integrated photonic platform, which enabled the control of nonHermicity, perform mathematical conversion to machine learning, respectively. 0D: Embedding individual symmetric or asymmetric meta-atoms in silicon micro- resonators provide the full control of non-Hermicity, which has been proved to coherently suppress the nanofabrication resulted backscattering [1]. 1D: The integrated metasystem performs analogue optical computing tasks, from simple Fourier transformation to spatial differentiations (1D+) [2]. Also, we have shown that asymmetric subwavelength design engineers the wave momentum space for broadband and power independent back reflection suppression. 2D: With lithographically defined inter-layer alignment, we demonstrate diffractive deep optical network on silicon photonic platform, towards broadband spatial pattern classification and hyperspectral imaging [3]. In addition to materials offered by the foundry, I will try to extend the scope of 'heterogeneous integration' for layered phase change materials for integrated photonic memory devices [4], and potential integration scheme with silicon photonics.
Biography: Tingyi Gu is an associate professor in the electrical engineering of University of Delaware. Her group works on foundry compatible silicon photonic meta-components for optical communication and sensing, with the focus on optoelectronic reconfigurability and high-speed operation. She served on 19 committees for optics and optoelectronics societies, including SPIE, CLEO, FiO and IPC. She received a B.S. from Shanghai Jiao Tong University, and M.S. and Ph.D. degrees from Columbia University, all in EE. She has held positions at Bell Labs, Princeton University and Hewlett Packard Labs.
Host: J Yang, H Wang, C Zhou, S Cronin, W Wu
More Information: Tingyi Gu_2024-02-09.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
Audiences: Everyone Is Invited
Contact: Marilyn Poplawski
-
ECE Virtual Seminar: Transdisciplinary Engineering: Reaching Beyond Engineering to Exploit Concepts From Other Disciplines
Mon, Feb 12, 2024 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Azad M. Madni, University Professor of Astronautics, Aerospace and Mechanical Engineering, USC Viterbi School of Engineering
Talk Title: Transdisciplinary Engineering: Reaching Beyond Engineering to Exploit Concepts From Other Disciplines
Abstract: This talk presents transdisciplinary engineering and how it is enabled by exploiting convergence of engineering with other disciplines. Specifically, it presents an overview of my research in this area, TRASEE™ educational paradigm, and the transformation of the Systems Architecting and Engineering Program using TRASEE. It focuses on storytelling in virtual worlds as an exemplar of exploiting convergence between engineering and entertainment/cinematic arts.
Biography: Azad Madni is a University Professor of Astronautics, Aerospace and Mechanical Engineering in the University of Southern California. The designation of University Professor honors USC’s most accomplished multidisciplinary faculty with significant achievements across multiple technical fields. He is the holder of the Northrop Grumman Fred O’Green Chair in Engineering, and the Executive Director of University of USC’s Systems Architecting and Engineering Program. He also holds a joint appointment in the Sonny Astani Department of Civil and Environmental Engineering, and courtesy appointments in the Rossier School of Education and Keck School of Medicine. He is the Founding Director of the Distributed Autonomy and Intelligent Systems Laboratory and is a faculty affiliate of USC’s Ginsberg Institute for Biomedical Therapeutics in the Keck School of Medicine. He is the founding director of the Ph.D. degree program in Systems Engineering in the Astronautics Department. He is also a Senior Fellow of the Loker Hydrocarbon Research Institute founded by Nobel Laureate, George Olah. He is a member of the London Digital Twin Research Centre. He is the founder and CEO of Intelligent Systems Technology, Inc., an award-winning hi-tech company specializing in model-based approaches for addressing scientific and societal problems of national and global significance. He is the Chief Systems Engineering Advisor to The Aerospace Corporation. He received his Ph.D., M.S., and B.S. degrees in Engineering from the University of California, Los Angeles. He is a graduate of AEA/Stanford Institute Executive Program for Technology Executives.
Host: Dr. Richard M. Leahy, leahy@usc.edu
Webcast: https://usc.zoom.us/j/91315597163?pwd=YjlrMlhGYnV4NEV4UkFiZXdETkZiQT09WebCast Link: https://usc.zoom.us/j/91315597163?pwd=YjlrMlhGYnV4NEV4UkFiZXdETkZiQT09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
-
Seminar
Fri, Feb 16, 2024 @ 03:30 PM - 04:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Ahmad Beirami, Google Research
Talk Title: Language Model Alignment: Theory & Practice
Series: AIF4S Seminar Series
Abstract: Generative language models have advanced to a level where they can effectively solve a variety of open-domain tasks with little task specific supervision. However, the generated content from these models may still not satisfy the preference of a human user. The goal of the alignment process is to remedy this issue by generating content from an aligned model that improves a reward (e.g., make the generation safer) but does not perturb much from the base model. A simple baseline for this task is best-of-N, where N responses are drawn from the base model, ranked based on a reward, and the highest ranking one is selected. More sophisticated techniques generally solve a KL-regularized reinforcement learning (RL) problem with the goal of maximizing expected reward subject to a KL divergence constraint between the aligned model and the base model. An alignment technique is preferred if its reward-KL tradeoff curve dominates other techniques. In this talk, we give an overview of language model alignment and give an understanding of known results in this space through simplified examples. We also present a new modular alignment technique, called controlled decoding, which solves the KL-regularized RL problem while keeping the base model frozen through learning a prefix scorer, offering inference-time configurability. Finally, we also shed light on the remarkable performance of best-of-N in terms of achieving competitive or even better reward-KL tradeoffs when compared to state-of-the-art alignment baselines.
Biography: Ahmad Beirami is a research scientist at Google Research, leading research efforts on building safe, helpful, and scalable generative language models. At Meta AI, he led research to power the next generation of virtual digital assistants with AR/VR capabilities through robust generative language modeling. At Electronic Arts, he led the AI agent research program for automated playtesting of video games and cooperative reinforcement learning. Before moving to industry in 2018, he held a joint postdoctoral fellow position at Harvard & MIT, focused on problems in the intersection of core machine learning and information theory. He is the recipient of the Sigma Xi Best PhD Thesis Award from Georgia Tech.
Host: Mahdi Soltanolkotabi
Webcast: https://usc.zoom.us/j/92673154833?pwd=Z1QwYk52RVhWSkRXRmhzTmRhUTU3UT09More Information: 14766.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 132
WebCast Link: https://usc.zoom.us/j/92673154833?pwd=Z1QwYk52RVhWSkRXRmhzTmRhUTU3UT09
Audiences: Everyone Is Invited
Contact: Gloria Halfacre
-
CSC/CommNetS-MHI Seminar: Yongduan Song
Tue, Feb 20, 2024 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Yongduan Song, Director, Research Institute for Artificial Intelligence | Chair Professor, School of Automation | Chongqing University
Talk Title: Several critical issues in neural network driven control design and analysis
Series: CSC/CommNetS-MHI Seminar Series
Abstract:
Neural networks (NN) and related learning algorithms are crucial components of artificial intelligence. The utilization of neural networks combined with learning algorithms for controller design has become a mainstream direction in the field of intelligent control. This talk will examine the typical NN-driven design approaches and expose several critical issues related to functionality and effectiveness of the NN-based control methods.
Biography:
Professor Yongduan Song is a Fellow of IEEE, Fellow of AAIA, Fellow of International Eurasian Academy of Sciences, and Fellow of Chinese Automation Association. He was one of the six Langley Distinguished Professors at National Institute of Aerospace (NIA), USA and registered professional engineer (USA). He is currently the dean of Research Institute of Artificial Intelligence at Chongqing University. Professor Song is the Editor-in-Chief of IEEE Transactions on Neural Networks and Learning Systems (TNNLS) and the founding Editor-in-Chief of the International Journal of Automation and Intelligence.
Host: Dr Petros Ioannou, ioannou@usc.edu
More Information: 2024.02.20 Seminar - Yongduan Song.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - EEB 132
Audiences: Everyone Is Invited
Contact: Miki Arlen
-
CSC/CommNetS-MHI Seminar: Milad Siami
Fri, Feb 23, 2024 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Milad Siami, Assistant Professor of Electrical and Computer Engineering | Northeastern University
Talk Title: Optimizing sparse interactions for control and sensing in complex networks
Series: CSC/CommNetS-MHI Seminar Series
Abstract:
This presentation introduces innovative strategies for enhancing control and sensing in large- scale complex networks, with a focus on minimizing resource usage to improve system performance. We address the challenge of non-submodular sensor scheduling in large-scale linear time-varying dynamics, tackling combinatorial, non-convex, NP-hard tasks. Beginning with a simple greedy algorithm, we present an approximation bound based on submodularity and curvature concepts, showing its superiority over existing methods. Shifting to discrete-time autonomous vehicle platoons, we employ graph- theoretic principles for state feedback laws, analyzing stability conditions based on underlying graph properties and update cycles. We explore H2-based robustness, demonstrating the impact of network density and update cycles on system performance. Specifically, we show that denser networks (i.e., networks with more communication links) might require faster agents (i.e., smaller update cycles) to outperform or achieve the same level of robustness as sparse networks (i.e., networks with fewer communication links). Practical examples and results from simulations and experiments, including work with Quanser's Qlabs and Qcars, validate the effectiveness of our approaches, emphasizing strategic sensor scheduling and robust design in autonomous vehicle platoons.
Biography:
Milad Siami is an Assistant Professor in the Department of Electrical and Computer Engineering at Northeastern University and a Core Faculty Member of the Institute for Experiential AI at the same institution. Prior to joining Northeastern, he served as a Postdoctoral Associate at the MIT Institute for Data, Systems, and Society. He earned his M.Sc. and Ph.D. degrees in Mechanical Engineering from Lehigh University and was a long- term visiting researcher at the Institute for Mathematics and Its Applications at the University of Minnesota. Additionally, he has experience as a Software Engineering Research Intern in the Modeling and Data Mining Group at Google Research NYC. Dr. Siami's research primarily focuses on the structural/graphical underpinnings of large-scale
dynamical networks and enhancing the reliability and security of AI-based autonomous systems. His specific areas of interest include distributed control systems, multi-robot systems, and autonomous networks. His current research is supported by grants from the National Science Foundation (NSF), the Department of Homeland Security (DHS), the Office of Naval Research (ONR), and the Army Research Laboratory (ARL).
Host: Dr. Mihailo Jovanovic
More Info: https://csc.usc.edu/seminars/2024Spring/siami.html
More Information: 2024.02.23 Seminar - Milad Siami.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - EEB 132
Audiences: Everyone Is Invited
Contact: Miki Arlen
Event Link: https://csc.usc.edu/seminars/2024Spring/siami.html
-
CSC/CommNetS-MHI Seminar: Milad Siami
Fri, Feb 23, 2024 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Milad Siami, Assistant Professor of Electrical and Computer Engineering | Northeastern University
Talk Title: Optimizing sparse interactions for control and sensing in complex networks
Series: CSC/CommNetS-MHI Seminar Series
Abstract:
This presentation introduces innovative strategies for enhancing control and sensing in large- scale complex networks, with a focus on minimizing resource usage to improve system performance. We address the challenge of non-submodular sensor scheduling in large-scale linear time-varying dynamics, tackling combinatorial, non-convex, NP-hard tasks. Beginning with a simple greedy algorithm, we present an approximation bound based on submodularity and curvature concepts, showing its superiority over existing methods. Shifting to discrete-time autonomous vehicle platoons, we employ graph- theoretic principles for state feedback laws, analyzing stability conditions based on underlying graph properties and update cycles. We explore H2-based robustness, demonstrating the impact of network density and update cycles on system performance. Specifically, we show that denser networks (i.e., networks with more communication links) might require faster agents (i.e., smaller update cycles) to outperform or achieve the same level of robustness as sparse networks (i.e., networks with fewer communication links). Practical examples and results from simulations and experiments, including work with Quanser's Qlabs and Qcars, validate the effectiveness of our approaches, emphasizing strategic sensor scheduling and robust design in autonomous vehicle platoons.
Biography:
Milad Siami is an Assistant Professor in the Department of Electrical and Computer Engineering at Northeastern University and a Core Faculty Member of the Institute for Experiential AI at the same institution. Prior to joining Northeastern, he served as a Postdoctoral Associate at the MIT Institute for Data, Systems, and Society. He earned his M.Sc. and Ph.D. degrees in Mechanical Engineering from Lehigh University and was a long- term visiting researcher at the Institute for Mathematics and Its Applications at the University of Minnesota. Additionally, he has experience as a Software Engineering Research Intern in the Modeling and Data Mining Group at Google Research NYC. Dr. Siami's research primarily focuses on the structural/graphical underpinnings of large-scale dynamical networks and enhancing the reliability and security of AI-based autonomous systems. His specific areas of interest include distributed control systems, multi-robot systems, and autonomous networks. His current research is supported by grants from the National Science Foundation (NSF), the Department of Homeland Security (DHS), the Office of Naval Research (ONR), and the Army Research Laboratory (ARL).
Host: Dr. Mihailo Jovanovic
More Info: https://csc.usc.edu/seminars/2024Spring/siami.html
More Information: 2024.02.23 Seminar - Milad Siami.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - EEB 132
Audiences: Everyone Is Invited
Contact: Miki Arlen
Event Link: https://csc.usc.edu/seminars/2024Spring/siami.html
-
CSC/CommNetS-MHI Seminar: Milad Siami
Fri, Feb 23, 2024 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Milad Siami, Assistant Professor of Electrical and Computer Engineering | Northeastern University
Talk Title: Optimizing sparse interactions for control and sensing in complex networks
Series: CSC/CommNetS-MHI Seminar Series
Abstract:
This presentation introduces innovative strategies for enhancing control and sensing in large- scale complex networks, with a focus on minimizing resource usage to improve system performance. We address the challenge of non-submodular sensor scheduling in large-scale linear time-varying dynamics, tackling combinatorial, non-convex, NP-hard tasks. Beginning with a simple greedy algorithm, we present an approximation bound based on submodularity and curvature concepts, showing its superiority over existing methods. Shifting to discrete-time autonomous vehicle platoons, we employ graph- theoretic principles for state feedback laws, analyzing stability conditions based on underlying graph properties and update cycles. We explore H2-based robustness, demonstrating the impact of network density and update cycles on system performance. Specifically, we show that denser networks (i.e., networks with more communication links) might require faster agents (i.e., smaller update cycles) to outperform or achieve the same level of robustness as sparse networks (i.e., networks with fewer communication links). Practical examples and results from simulations and experiments, including work with Quanser's Qlabs and Qcars, validate the effectiveness of our approaches, emphasizing strategic sensor scheduling and robust design in autonomous vehicle platoons.
Biography:
Milad Siami is an Assistant Professor in the Department of Electrical and Computer Engineering at Northeastern University and a Core Faculty Member of the Institute for Experiential AI at the same institution. Prior to joining Northeastern, he served as a Postdoctoral Associate at the MIT Institute for Data, Systems, and Society. He earned his M.Sc. and Ph.D. degrees in Mechanical Engineering from Lehigh University and was a long- term visiting researcher at the Institute for Mathematics and Its Applications at the University of Minnesota. Additionally, he has experience as a Software Engineering Research Intern in the Modeling and Data Mining Group at Google Research NYC. Dr. Siami's research primarily focuses on the structural/graphical underpinnings of large-scale dynamical networks and enhancing the reliability and security of AI-based autonomous systems. His specific areas of interest include distributed control systems, multi-robot systems, and autonomous networks. His current research is supported by grants from the National Science Foundation (NSF), the Department of Homeland Security (DHS), the Office of Naval Research (ONR), and the Army Research Laboratory (ARL).
Host: Dr. Mihailo Jovanovic
More Information: 2024.02.23 Seminar - Milad Siami.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - EEB 132
Audiences: Everyone Is Invited
Contact: Miki Arlen
-
CSC/CommNetS-MHI Seminar: Milad Siami
Fri, Feb 23, 2024 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Milad Siami, Assistant Professor of Electrical and Computer Engineering | Northeastern University
Talk Title: Optimizing sparse interactions for control and sensing in complex networks
Series: CSC/CommNetS-MHI Seminar Series
Abstract:
This presentation introduces innovative strategies for enhancing control and sensing in large- scale complex networks, with a focus on minimizing resource usage to improve system performance. We address the challenge of non-submodular sensor scheduling in large-scale linear time-varying dynamics, tackling combinatorial, non-convex, NP-hard tasks. Beginning with a simple greedy algorithm, we present an approximation bound based on submodularity and curvature concepts, showing its superiority over existing methods. Shifting to discrete-time autonomous vehicle platoons, we employ graph- theoretic principles for state feedback laws, analyzing stability conditions based on underlying graph properties and update cycles. We explore H2-based robustness, demonstrating the impact of network density and update cycles on system performance. Specifically, we show that denser networks (i.e., networks with more communication links) might require faster agents (i.e., smaller update cycles) to outperform or achieve the same level of robustness as sparse networks (i.e., networks with fewer communication links). Practical examples and results from simulations and experiments, including work with Quanser's Qlabs and Qcars, validate the effectiveness of our approaches, emphasizing strategic sensor scheduling and robust design in autonomous vehicle platoons.
Biography:
Milad Siami is an Assistant Professor in the Department of Electrical and Computer Engineering at Northeastern University and a Core Faculty Member of the Institute for Experiential AI at the same institution. Prior to joining Northeastern, he served as a Postdoctoral Associate at the MIT Institute for Data, Systems, and Society. He earned his M.Sc. and Ph.D. degrees in Mechanical Engineering from Lehigh University and was a long- term visiting researcher at the Institute for Mathematics and Its Applications at the University of Minnesota. Additionally, he has experience as a Software Engineering Research Intern in the Modeling and Data Mining Group at Google Research NYC. Dr. Siami's research primarily focuses on the structural/graphical underpinnings of large-scale dynamical networks and enhancing the reliability and security of AI-based autonomous systems. His specific areas of interest include distributed control systems, multi-robot systems, and autonomous networks. His current research is supported by grants from the National Science Foundation (NSF), the Department of Homeland Security (DHS), the Office of Naval Research (ONR), and the Army Research Laboratory (ARL).
Host: Dr. Mihailo Jovanovic, mihailo@usc.edu
More Info: https://csc.usc.edu/seminars/2024Spring/siami.html
More Information: 2024.02.23 Seminar - Milad Siami.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - EEB 132
Audiences: Everyone Is Invited
Contact: Miki Arlen
Event Link: https://csc.usc.edu/seminars/2024Spring/siami.html
-
CSC/CommNetS-MHI Seminar: Milad Siami
Fri, Feb 23, 2024 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Milad Siami, Assistant Professor of Electrical and Computer Engineering | Northeastern University
Talk Title: Optimizing sparse interactions for control and sensing in complex networks
Series: CSC/CommNetS-MHI Seminar Series
Abstract:
This presentation introduces innovative strategies for enhancing control and sensing in large- scale complex networks, with a focus on minimizing resource usage to improve system performance. We address the challenge of non-submodular sensor scheduling in large-scale linear time-varying dynamics, tackling combinatorial, non-convex, NP-hard tasks. Beginning with a simple greedy algorithm, we present an approximation bound based on submodularity and curvature concepts, showing its superiority over existing methods. Shifting to discrete-time autonomous vehicle platoons, we employ graph- theoretic principles for state feedback laws, analyzing stability conditions based on underlying graph properties and update cycles. We explore H2-based robustness, demonstrating the impact of network density and update cycles on system performance. Specifically, we show that denser networks (i.e., networks with more communication links) might require faster agents (i.e., smaller update cycles) to outperform or achieve the same level of robustness as sparse networks (i.e., networks with fewer communication links). Practical examples and results from simulations and experiments, including work with Quanser's Qlabs and Qcars, validate the effectiveness of our approaches, emphasizing strategic sensor scheduling and robust design in autonomous vehicle platoons.
Biography:
Milad Siami is an Assistant Professor in the Department of Electrical and Computer Engineering at Northeastern University and a Core Faculty Member of the Institute for Experiential AI at the same institution. Prior to joining Northeastern, he served as a Postdoctoral Associate at the MIT Institute for Data, Systems, and Society. He earned his M.Sc. and Ph.D. degrees in Mechanical Engineering from Lehigh University and was a long- term visiting researcher at the Institute for Mathematics and Its Applications at the University of Minnesota. Additionally, he has experience as a Software Engineering Research Intern in the Modeling and Data Mining Group at Google Research NYC. Dr. Siami's research primarily focuses on the structural/graphical underpinnings of large-scale dynamical networks and enhancing the reliability and security of AI-based autonomous systems. His specific areas of interest include distributed control systems, multi-robot systems, and autonomous networks. His current research is supported by grants from the National Science Foundation (NSF), the Department of Homeland Security (DHS), the Office of Naval Research (ONR), and the Army Research Laboratory (ARL).
Host: Dr. Mihailo Jovanovic, mihailo@usc.edu
More Info: https://csc.usc.edu/seminars/2024Spring/siami.html
More Information: 2024.02.23 Seminar - Milad Siami.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - EEB 132
Audiences: Everyone Is Invited
Contact: Miki Arlen
Event Link: https://csc.usc.edu/seminars/2024Spring/siami.html
-
ECE Seminar
Fri, Feb 23, 2024 @ 03:30 PM - 04:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Jorge F. Silva, PhD, Universidad de Chile
Talk Title: Information Theoretic Measures for Representation Learning
Abstract: Information-theoretic measures have been widely adopted for machine learning (ML) feature design. Inspired by this, we look at the relationship between information loss in the Shannon sense and the operation loss in the minimum probability of error (MPE) sense when considering a family of lossy representations (or encoders). In this talk, we introduce a series of results that show how adequate the adoption of mutual information (MI) is for predicting the operational quality of a representation in classification. Our findings support the observation that selecting/designing representations that capture informational sufficiency (IS) is appropriate for learning. However, we also show that this selection is rather conservative if the intended goal is achieving MPE in classification. We conclude by discussing the capacity of the information bottleneck (IB) method to achieve lossless prediction and the expressive power of digital encoders in ML.
Biography: Information-theoretic measures have been widely adopted for machine learning (ML) feature design. Inspired by this, we look at the relationship between information loss in the Shannon sense and the operation loss in the minimum probability of error (MPE) sense when considering a family of lossy representations (or encoders). In this talk, we introduce a series of results that show how adequate the adoption of mutual information (MI) is for predicting the operational quality of a representation in classification. Our findings support the observation that selecting/designing representations that capture informational sufficiency (IS) is appropriate for learning. However, we also show that this selection is rather conservative if the intended goal is achieving MPE in classification. We conclude by discussing the capacity of the information bottleneck (IB) method to achieve lossless prediction and the expressive power of digital encoders in ML.
Host: Dr. Eduardo Pavez
More Information: Jorge Silva Seminar 2.23.24.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 132
Audiences: Everyone Is Invited
Contact: Gloria Halfacre
-
CSC/CommNetS-MHI Seminar: Ingvar Ziemann
Mon, Feb 26, 2024 @ 02:00 PM - 03:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Ingvar Ziemann, Postdoctoral Researcher | University of Pennsylvania
Talk Title: Sharp rates in dependent learning theory
Series: CSC/CommNetS-MHI Seminar Series
Abstract: In this talk I discuss some recent advances in supervised learning with dependent data. In particular, the emphasis of this talk is to provide an instance-optimal understanding of learning with dependent data for the square loss function. The approach I present yields rates that match and extend known asymptotics even without any realizability assumption. This stands in stark contrast to typical non-asymptotic results which exhibit variance proxies that are deflated multiplicatively by the mixing time of the underlying data-generating process. Indeed, our results instead scale additively with the mixing time and are thereby only affected by second order statistics in the leading term. The key to obtaining this scaling is the introduction of the notion of a weakly sub-Gaussian class, which allows us to invoke mixed tail generic chaining. This notion is general enough to nearly all cover smooth hypothesis classes and a wide range of parametric classes. As a motivating example, I will also discuss our recent work on multi-task learning. Even when the problem itself is realizable, the analysis of a natural “two-stage” estimator decomposes into two supervised learning problems: one which is realizable, and one which is not. In this setting, we demonstrate how our refined understanding of supervised learning with dependent data can be applied to extend and sharpen existing guarantees for iid multi-task learning.
Biography: Ingvar Ziemann is a postdoctoral researcher at the University of Pennsylvania. He received his PhD in November 2022 from the Division of Decision and Control Systems at The Royal Institute of Technology (KTH) under the supervision of Henrik Sandberg. His research is centered on using statistical and information theoretic tools to study learning-enabled control methods, with a current interest in studying how learning algorithms generalize in the context of dynamical systems. Prior to starting his Ph.D., he obtained two sets of Master's and Bachelor's degrees in Mathematics (SU/KTH) and in Economics and Finance (SSE). Ingvar is the recipient of a Swedish Research Council International Postdoc Grant, the 2022 IEEE Conference on Decision and Control Best Student Paper Award, and the 2017 Stockholm Mathematics Center Excellent Master Thesis Award.
Host: Dr. Lars Lindemann
More Info: https://csc.usc.edu/seminars/2024Spring/ziemann.html
More Information: 2024.02.26 CSC Seminar - Ingvar Ziemann.pdf
Location: Hughes Aircraft Electrical Engineering Center (EEB) - EEB 248
Audiences: Everyone Is Invited
Contact: Miki Arlen
Event Link: https://csc.usc.edu/seminars/2024Spring/ziemann.html
-
ECE Seminar: Dr. Giacomo Nannicini
Tue, Feb 27, 2024 @ 11:00 AM - 12:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Giacomo Nannicini, Associate Professor, Epstein Dept of ISE, USC Viterbi School of Engineering
Talk Title: Convex Optimization Algorithms on Quantum Computers
Abstract: Optimization is often mentioned as one of the main application areas for quantum computers, but is this claim backed up by theoretical evidence? In this talk we provide a gentle overview of recent advances in quantum optimization, with an emphasis on algorithms and subroutines for convex optimization problems that lead to rigorous asymptotic speedups. The main results of this talk are a faster classical algorithm for the semidefinite relaxation of the MaxCut problem, an even faster quantum algorithm for the same problem, and a new idea for linear optimization on quantum computers.
Biography: Giacomo Nannicini is an associate professor in the Industrial & Systems Engineering department at the University of Southern California, which he joined in 2022. Prior to that, he was a research staff member in the quantum algorithms group at the IBM T. J. Watson Research Center, and an assistant professor in the Engineering Systems and Design pillar at the Singapore University of Technology and Design. His main research interest is optimization broadly defined and its applications. Giacomo received several awards, including the 2021 Beale--Orchard-Hays prize, the 2015 Robert Faure prize, and the 2012 Glover-Klingman prize.
Host: Dr. Richard M. Leahy, leahy@usc.edu
Webcast: https://usc.zoom.us/j/95762332255?pwd=NitkT2p5c1kvWWp0a0JuUUVNZTRudz09Location: Hughes Aircraft Electrical Engineering Center (EEB) - 248
WebCast Link: https://usc.zoom.us/j/95762332255?pwd=NitkT2p5c1kvWWp0a0JuUUVNZTRudz09
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
-
ECE-EP Seminar - Aziza Suleymanzade, Tuesday, Feb. 27th at 2pm via Zoom
Tue, Feb 27, 2024 @ 02:00 PM - 03:30 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Aziza Suleymanzade, Harvard University
Talk Title: Building quantum networks: from solid-state defects and Rydberg atoms in cavities to a new scientific frontier with hybrid quantum systems
Series: ECE-EP Seminar
Abstract: The experimental development of quantum networks marks a significant scientific milestone, poised to enable secure quantum communication, distributed quantum computing, and entanglement-enhanced nonlocal sensing. In this talk, I will discuss the recent advancements in the field along with the outstanding challenges through my work on two different platforms: Silicon Vacancy defects in diamond nanophotonic cavities and Rydberg atoms coupled to hybrid cavities. First, I will present our recent results on distributing entanglement across a two-node network with on-chip solid-state defects in cavities which we built at Harvard. We demonstrated high-fidelity entanglement between communication and memory qubits and showed long-distance entanglement over the 35 km of deployed fiber in the Cambridge/Boston area. Second, I will describe our work at the University of Chicago on using Rydberg atoms as transducers of quantum information between optical and microwave photons, with the goal of integrating Rydberg platforms with superconducting circuits and paving the way for advanced quantum network architectures. The talk will conclude with a perspective on the potential of this hybrid platform approach in constructing quantum networks, highlighting the uncharted scientific and technological opportunities it could unlock.
Biography: Aziza is a postdoc at Harvard in the group of Mikhail Lukin. She did her PhD at the University of Chicago in groups of Jon Simon and David Schuster, working on the transduction of single optical to millimeter wave photons using Rydberg atoms in cavities. Aziza got a Bachelor's degree from Harvard University and an MPhil from the University of Cambridge, where she built an experiment for generating potassium-39 BEC in a uniform box potential.
Host: ECE-EP
More Info: https://usc.zoom.us/j/96689616375?pwd=bGJ0dXZZUEdxTjN3bHFlL3ZnVWdVUT09
More Information: Aziza Suleymanzade Seminar Announcement.pdf
Audiences: Everyone Is Invited
Contact: Marilyn Poplawski
Event Link: https://usc.zoom.us/j/96689616375?pwd=bGJ0dXZZUEdxTjN3bHFlL3ZnVWdVUT09
-
2024 Viterbi Keynote Lecture
Thu, Feb 29, 2024 @ 03:00 PM - 04:00 PM
Ming Hsieh Department of Electrical and Computer Engineering
Conferences, Lectures, & Seminars
Speaker: Dr. Arogyaswami Paulraj, Emeritus Professor, Stanford University
Talk Title: Big Ideas in Mobile Wireless Technology: Many are Called, but Only Some are Chosen
Series: Viterbi Lecture
Abstract: This talk takes a panoramic view of the evolution of mobile wireless technology from 2G to 5G. The research community has put forth several significant ideas, but only some (as of yet) have actually made it into mobile standards. This talk takes a somewhat simplistic (but hopefully accessible) view of these ideas and outlines the complex tradeoffs that pick winners and losers.
Biography: Paulraj is an Emeritus Professor at Stanford University and a pioneer of MIMO (Multiple Input, Multiple Output) wireless, the key technology adopted in all modern wireless systems. Paulraj served for 25 years with the Indian Navy, leading programs in ASW Naval sonar systems for a decade and, for shorter periods, other major Indian national initiatives in AI, high-speed computing, and combat jet aircraft. He received a Ph.D. from the Indian Institute of Technology, New Delhi, India, in 1973. After prematurely retiring from the Navy in 1991, Paulraj joined Stanford University as a research associate. Paulraj founded Iospan Wireless Inc., which pioneered MIMO-OFDMA wireless technology. He co-founded Beceem Communications Inc., which became the leader in 4G-WiMAX chip sets. And later, he founded Rasa Networks for AI-based WiFi network analytics. These companies were acquired by Intel, Broadcom, and HPE, respectively. Paulraj's recognitions include the 2023 IET Faraday Medal, the 2014 Marconi Prize, the 2011 IEEE Alexander Graham Bell Medal, the 2018 Induction into the US Patent Office’s National Inventors Hall of Fame, and the 2022 Induction into the Wireless History Foundation’s Hall of Fame. He is a member of ten national academies spanning engineering, the sciences, and the arts. His recognitions also include the Friendship Award from the Government of PR China and the Padma Bhushan from the Government of India.
Host: Dr. Richard M. Leahy, leahy@usc.edu
Location: Hughes Aircraft Electrical Engineering Center (EEB) - 132
Audiences: Everyone Is Invited
Contact: Mayumi Thrasher
