Technical Program

New Frontiers in Networked Dynamical Systems:

Assured Learning, Communication and Control

Technical program schedule

Clark Hall [Forum 1101], Univ. of Maryland College Park Campus

Friday, Oct. 13 (morning)

7:30-8:15am: Registration, breakfast

8:15-8:30am: Opening remarks

Robert M. Briber, A. J. Clark School of Engineering, University of Maryland 

Ankur Srivastava, Director, Institute for Systems Research 

8:30-9:00am: Keynote Talk (Session Chair: Mingyan Liu)

Softly shaping behavior

Tamer Başar, University of Illinois Urbana-Champaign

9:00-9:20am: Coffee break

9:20-11:00am: Short Talks: Multi-agent Systems (Session Chair: John Baillieul)

(9:20-9:40am) Spatiotemporal logic control for leader-follower multi-agent systems

Dimos Dimarogonas, KTH Royal Institute of Technology, Sweden

(9:40-10:00am) Information design under uncertainty

Munther Dahleh, MIT

(10-10:20am) Swarm-based gradient descent methods for non-convex optimization

Eitan Tadmor, University of Maryland

(10:20-10:40am) Opinion dynamics under social pressure in arbitrary networks

Ali Jadbabaie, MIT

(10:40-11:00am) Enabling new paradigms for networked dynamical systems through 6G

Andrea Goldsmith, Princeton University

11:00-11:10am: Short break

11:10am-12:10pm: Panel 1: Duel or Duet: The Fate of Control Theory in the Era of Machine Learning and AI

Xiaobo Tan (moderator), Michigan State University

Eyad Abed, University of Maryland, College Park

John Baillieul, Boston University

Maryam Fazel, University of Washington  

Na Li, Harvard University

12:10-1:50pm: Lunch Break and Keynote (Session Chair: Nikolaos Sidiropoulos)

(12:45-1:15pm) Retrospective and state of the Institute for Systems Research (ISR)

Ankur Srivastava, University of Maryland

Friday, Oct. 13 (afternoon)

1:50-2:50pm: Panel 2: How AI is Shaping Future Systems Engineering

Hongjun “Jason” Li (moderator), Trusted Science and Technology

John S. Baras, University of Maryland

Albert Benveniste, Inria, France

Mauricio Castillo-Effen, Lockheed Martin

John D. Kenyon, Hughes Network Systems (retired)

2:50-3:20pm: Coffee break

3:20-3:50pm: Keynote Talk (Session Chair: Armand Makowski)

Shared information

Prakash Narayan, University of Maryland

3:50-5:30pm: Short Talks: Learning Theory and Learning-based Control (Session Chair: Radha Poovendran)

(3:50-4:10pm) Asymptotic learning in overparameterized models

Anant Sahai, University of California, Berkeley

(4:10-4:30pm) Policy optimization methods in control

Maryam Fazel, University of Washington

(4:30-4:50pm) Higher-order learning vs Nash equilibrium

Jeff Shamma, University of Illinois Urbana-Champaign

(4:50-5:10pm) Representation-based nonlinear stochastic control and learning via spectral embedding

Na Li, Harvard University

(5:10-5:30pm) Learning dynamics of overparametrized networks

René Vidal, University of Pennsylvania

5:30-5:35pm: Closing remarks for Day 1

Saturday, Oct. 14 (morning)

7:45-8:15am: Registration, breakfast

8:15-8:20am: Opening remarks

8:20-8:50am: Keynote Talk (Session Chair: Miroslav Krstic)

The Reward-biased method: Exploration vs. exploitation from adaptive control to reinforcement learning

P. R. Kumar, Texas A&M University

8:50-9:00am: Short break

9:00-10:40am: Short Talks: Stochastic and Distributed Systems and Control (Session Chair: Jeff Shamma)

(9:00-9:20am) Stochastic control and thermodynamics in finite time

Tryphon T. Georgiou, University of California, Irvine

(9:20-9:40am) Stochastic safety noncooperative games

Miroslav Krstic, University of California, San Diego

(9:40-10:00am) Operator-valued kernels and control of infinite dimensional dynamic systems

Alain Bensoussan, University of Texas, Dallas

(10:00-10:20am) A variational approach to sampling and path estimation in diffusion processes

Maxim Raginsky, University of Illinois Urbana-Champaign

(10:20-10:40am) Stochastic systems with Rosenblatt noise and applications

Tyrone Duncan, University of Kansas

10:40-10:50am: Short break

10:50am-12:30pm: Short Talks: Learning, Control, and Security in Communications and Networks (Session Chair: Nikolaos Sidiropoulos)

(10:50-11:10am) Federated learning for cooperative control over additive channel

Tara Javidi, University of California, San Diego

(11:10-11:30am) Together we can: DDoS attack detection via privacy-aware federated learning and collaborative mitigation in multi-domain cyber infrastructures

Vasilis Maglaris, National Technical University of Athens (NTUA), Greece

(11:30-11:50am) How much is a bit worth?

Massimo Franceschetti, University of California, San Diego

(11:50am-12:10pm) Securing location-based mobile computing

Panos Papadimitratos, KTH Royal Institute of Technology, Sweden

(12:10-12:30pm) Randomization-based defenses against membership inference attacks and backdoor attacks in deep neural networks

Radha Poovendran, University of Washington

12:30-2:00pm: Lunch Break and Keynote (Session Chair: Hongjun “Jason” Li)

(1:15-1:45pm) Self organization: From neural networks to networked systems

John S. Baras, University of Maryland

Saturday, Oct. 14 (afternoon)

2:00-2:30pm: Keynote Talk (Session Chair: Alvaro Cardenas)

When things get big: Using hierarchy to gain analytical and computational advantage in multi-agent systems

Mingyan Liu, University of Michigan

2:30-4:10pm: Short Talks: Modeling and Analysis of Networked Complex Systems (Session Chair: Eyad Abed)

(2:30-2:50pm) Large physics dominated cyberphysical systems modeling: DAEs as needed solution but scientific nightmare

Albert Benveniste, Inria, France

(2:50-3:10pm) Predicting oscillations in relay feedback systems

Maben Rabi, Østfold University College, Norway

(3:10-3:30pm) Autonomy, GNNs, and spectrum

Brian Sadler, Army Research Laboratory

(3:30-3:50pm) A quantum walk model for emotion transmission in serial reproduction of narratives (via Zoom)

Jerome Busemeyer, Indiana University

(3:50-4:10pm) On the value of data for learning-based control

Sandra Hirche, Technical University of Munich, Germany

4:10-4:30pm: Coffee break

4:30-5:30pm: Panel 3: The Era of Generative AI: Assured Human-cyber-physical Systems

Nikolaos Sidiropoulos (moderator), University of Virginia

Rance Cleaveland, University of Maryland

Panos Papadimitratos, KTH Royal Institute of Technology, Sweden

Radha Poovendran, University of Washington

René Vidal, University of Pennsylvania

5:30-5:45pm: Closing remarks

John S. Baras, University of Maryland