Events for February 23, 2024
-
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