Uncategorized

CESG is proud to announce our Fall 2022 graduates who are entering the world of industry and academia after unique times of online learning, masked classes, and other firsts that Texas A&M implemented until we reached the other side of such pandemic safety concerns.  We hope they will take with them the knowledge that:

• they can reach their goals despite a changing environment,
• the content studied gives them a strong foothold in their careers,
• personal connections from classes, clubs, and faculty can be a long-term resource to them,
• we are thankful to have had them in the Computer Engineering Systems Group of ECE at TAMU, and
• the 6 Aggie Core Values are traditions of A&M that, when practiced, will serve them throughout their lives.

Congratulations Graduates!

Doctor of Philosophy
Dr. Caio Davi (Advisor: Braga-Neto)
Dr. Luke McHale (Advisors: Gratz & Sprintson)
Dr. Jingqing Wang (Advisor: Zhang)

Master of Engineering
Florence Johnson
Rishya Sankar Kumaran
Nhat Do Minh Nguyen
Jaspal Singh Nijjar
Nonetta Marie Pierre
Lingaraju Elavarthi Vasuraju
Wei Wei
Sai Suma Yallapragda

Master of Science
Kunal Bharathi
Sourav Dokania
Haroon Faisal
Sayan Ganguly
Chien-Peng Huang
Karanam Hemachandran
Ravi Pranjal
Ruiyu Qu
Niranjan Raghuraman
Rohith Ramanujam Kumar
Santwana
Alexander J. Staggs
Ting Wei Su
Sowmya Sree Thota

Commencement: Saturday, 12/17/22 @ 9 a.m. in Reed Arena

Friday, November 18, 2022
10:20 – 11:10 a.m. (CST)
Zoom (Links and PW in syllabus or email)

Dr. Sanjay Shakkottai
Professor, Department of Electrical and Computer Engineering
University of Texas at Austin

Title: “The Power of Adaptivity in Representation Learning: from Meta-Learning to Federated Learning”

Talking Points

• Algorithms for multi-task learning that learn representation
• Understanding the training dynamics of meta-learning and federated averaging with fine tuning

Abstract
A central problem in machine learning is as follows: How should we train models using data generated from a collection of clients/environments, if we know that these models will be deployed in a new and unseen environment? In the setting of few-shot learning, two prominent approaches are: (a) develop a modeling framework that is “primed” to adapt, such as Model Adaptive Meta Learning (MAML), or (b) develop a common model using federated learning (such as FedAvg), and then fine tune the model for the deployment environment. We study both these approaches in the multi-task linear representation setting. We show that the reason behind generalizability of the models in new environments trained through either of these approaches is that the dynamics of training induces the models to evolve toward the common data representation among the clients’ tasks. In both cases, the structure of the bi-level update at each iteration (an inner and outer update with MAML, and a local and global update with FedAvg) holds the key — the diversity among client data distributions are exploited via inner/local updates, and induces the outer/global updates to bring the representation closer to the ground-truth. In both these settings, these are the first results that formally show representation learning, and derive exponentially fast convergence to the ground-truth representation. Based on joint work with Liam Collins, Hamed Hassani, Aryan Mokhtari, and Sewoong Oh. Papers: https://arxiv.org/abs/2202.03483 , https://arxiv.org/abs/2205.13692

Biography
Dr. Sanjay Shakkottaireceived his Ph.D. from the ECE Department at the University of Illinois at Urbana-Champaign in 2002. He is with The University of Texas at Austin, where he is a Professor in the Department of Electrical and Computer Engineering, and holds the Cockrell Family Chair in Engineering #15. He received the NSF CAREER award in 2004 and was elected as an IEEE Fellow in 2014. He was a co-recipient of the IEEE Communications Society William R. Bennett Prize in 2021. He is currently the Editor in Chief of IEEE/ACM Transactions on Networking. His research interests lie at the intersection of algorithms for resource allocation, statistical learning and networks, with applications to wireless communication networks and online platforms.

Webpage to learn more about Dr. Shakkottai: HERE

More on CESG Seminars: HERE

Please join on Friday, 11/18/22 at 10:20 a.m. in ETB 1020.

Friday, November 11, 2022
10:20 – 11:10 a.m. (CST)
Virtual via Zoom: https://tamu.zoom.us/j/93347193479 (password in emails or syllabus)

Dr.  Mohammad Ghavamzadeh
Senior Staff Research Scientist
Google

Title: “Mitigating the Risk Associated with Epistemic and Aleatory Uncertainties in MDPs”

Abstract
Prior work on safe reinforcement learning (RL) has studied risk-aversion to randomness in dynamics (aleatory) and to model uncertainty (epistemic) in isolation. We propose and analyze a new framework to jointly model the risk associated with epistemic and aleatory uncertainties in finite-horizon and discounted infinite-horizon MDPs. We call this framework that combines risk-averse and soft-robust methods RASR. We show that when the risk-aversion is defined using either the entropic value-at-risk (EVaR) or the entropic risk measure (ERM), the optimal policy in RASR can be computed efficiently using a new dynamic program formulation with a time-dependent risk level. As a result, the optimal risk-averse policies are deterministic but time-dependent, even in the infinite-horizon discounted setting. We also show that particular RASR objectives reduce to risk-averse RL with mean posterior transition probabilities. Our empirical results show that our new algorithms consistently mitigate uncertainty as measured by EVaR and other standard risk measures.

Biography 
Dr. Mohammad Ghavamzadeh received a Ph.D. degree from UMass Amherst in 2005. He was a postdoctoral fellow at UAlberta from 2005 to 2008. He was a permanent researcher at INRIA from 2008 to 2013. He was the recipient of the “INRIA award for scientific excellence” in 2011, and obtained his Habilitation in 2014. Since 2013, he has been a senior researcher at Adobe and FAIR, and now a senior staff research scientist at Google. He has published over 100 refereed papers in major machine learning, AI, and control journals and conferences. He has co-chaired more than 10 workshops and tutorials at NeurIPS, ICML, and AAAI. His research has been mainly focused on the areas of reinforcement learning, bandit algorithms, and recommendation systems.

More information on Dr. Ghavamzadeh can be found at
https://mohammadghavamzadeh.github.io/
https://scholar.google.ca/citations?user=Bo-wyrkAAAAJ&hl=en

More info. on past and future CESG Seminars at CESG Seminars (tamu.edu)

* Friday, 11/11/22 at 10:20 a.m. via Zoom *

Friday, November 4, 2022
10:20 – 11:10 a.m. (CST)
ETB 1020 – **In-person** (Zoom option; Links and PW in syllabus or email)

Dr. Alan Kuhnle
Assistant Professor, Computer Science and Engineering
Texas A&M University

Title: “Scalable and Learned Algorithms for Discrete Optimization”

Talking Points

• Linear-time algorithms for subset selection problems
• RL for learning local search algorithm

Abstract
In this talk, I present the work of the Optimization and Learning Systems Lab on the design of scalable algorithms for optimization problems on big data. In particular, I describe our work on linear-time, parallelizable algorithms for combinatorial optimization problems arising from online social networks. Finally, I give an overview of future directions of the lab, which include augmenting algorithms with learned components to improve practical performance; optimization with incomplete information; and submodular planning.

Biography
Dr. Alan Kuhnle is an Assistant Professor of Computer Science & Engineering at Texas A&M University, where he directs the Optimization and Learning Systems Lab. His work focuses on the design and analysis of scalable algorithms for ubiquitous combinatorial optimization problems arising in data science applications, such as vehicle routing and marketing on social networks. He is the recipient of the First Year Assistant Professor Award at Florida State University in 2020 and his work has led to 34 publications in leading academic journals and conferences. He has served on the program committee of machine learning conferences and is Associate Editor of Journal of Combinatorial Optimization.

Recent TAMU article on Dr. Kuhnle: HERE

More on CESG Seminars: HERE

Please join on Friday, 11/4/22 at 10:20 a.m. in ETB 1020.

Friday, October 28, 2022
10:20 – 11:10 a.m. (CST)
Virtual via Zoom: https://tamu.zoom.us/j/93347193479 (password in emails or syllabus)

Dr.  Dileep Kalathil
Assistant Professor in the Dept. of Electrical and Computer Engineering
Texas A&M University

Title: “Reinforcement Learning with Robustness and Safety Guarantees”

Talking Points
*How do we develop reinforcement learning algorithms that can overcome the simulation-to-reality gap?
*How do we develop reinforcement learning algorithms that maintain safety constraints during learning?

Abstract
Reinforcement Learning (RL) is the class of machine learning that addresses the problem of learning to control unknown dynamical systems. RL has achieved remarkable success recently in applications like playing games and robotics. However, most of these successes are limited to very structured or simulated environments. When applied to real-world systems, RL algorithms face two fundamental sources of fragility. First, the real-world system parameters can be very different from that of the nominal values used for training RL algorithms. Second, the control policy for any real-world system is required to maintain some necessary safety criteria to avoid undesirable outcomes. Most deep RL algorithms overlook these fundamental challenges which often results in learned policies that perform poorly in the real-world settings. In this talk, I will present two approaches to overcome these challenges. First, I will present an RL algorithm that is robust against the parameter mismatches between the simulation system and the real-world system. Second, I will discuss a safe RL algorithm to learn policies such that the frequency of visiting undesirable states and expensive actions satisfies the safety constraints. I will also briefly discuss some practical challenges due to the sparse reward feedback and the need for rapid real-time adaptation in real-world systems, and the approaches to overcome these challenges.
Robust RL papers: R-P1, R-P2, R-P3
Safe RL papers: S-P1, S-P2, S-P3, S-P4

Biography 
Dr. Dileep Kalathil  is an Assistant Professor in the Department of Electrical and Computer Engineering here at Texas A&M University (TAMU). His main research area is reinforcement learning theory and algorithms, and their applications in communication networks and power systems. Before joining TAMU, he was a postdoctoral researcher in the EECS department at UC Berkeley. He received his Ph.D. from University of Southern California (USC) in 2014, where he won the best Ph.D. Dissertation Prize in the Department of Electrical Engineering. He received his M. Tech. from IIT Madras, where he won the award for the best academic performance in the Electrical Engineering Department. He received the NSF CRII Award in 2019 and the NSF CAREER award in 2021. He is a senior member of IEEE.

More information on Dr. Kalathil can be found HERE.

More info. on past and future CESG Seminars at CESG Seminars (tamu.edu)

* Friday, 10/28/22 at 10:20 a.m. via Zoom *

Friday, October 7, 2022
10:20 – 11:10 a.m. (CST)
ETB 1020 – **In-person** (Zoom option; Links and PW in syllabus or email)

Dr. George T. Ligler
Professor, Multidisciplinary Engineering
Texas A&M University

Title: “Automatic Dependent Surveillance—Broadcast (ADS-B): A Major Aviation System’s Conception, Development, and Operational Implementation”

Talking Points

• In order to be successful, computer systems engineers need to learn to work with the full set of project stakeholders.
• The “best” technical solution to a problem is often not the one employed or that should be employed–broaden your view of he “trade space”!
• Computer system engineers have the opportunity to constantly learn while being instrumental in “getting things done”.
• We have a worldwide shortage of interdisciplinary computer systems engineers.

Abstract
ADS-B provides the largest improvement in air traffic surveillance since the advent of radar in the 1940’s. This presentation discusses the 27-year journey of the ADS-B System from concept development to operational implementation on over 100,000 aircraft in the U.S. National Airspace System alone. The role of interdisciplinary computer system engineering in the development and implementation of this high-impact system is highlighted.

Biography
Dr. George T. Ligler is a Professor and Dean’s Excellence Chair of Multidisciplinary Engineering at Texas A&M University. He is also the proprietor of GTL Associates, a consultancy which has provided systems integration/engineering and product management services in multiple fields to over 40 clients on three continents. An elected member of the U.S. National Academy of Engineering, Dr. Ligler is a member and past Chair of the Academy’s Section 12, Special Fields and Interdisciplinary Engineering.  With regard to aviation, he has served on one or more aviation standards development committees since 1992 and since 2005 has been the Co-Chair of RTCA Special Committee 159, Navigation Equipment Using the Global Navigation Satellite System (GNSS). He has been recognized with RTCA’s highest award, the Achievement Award, in 2006 and 2017 and was a co-recipient of the Air Traffic Control Association’s 2015 Chairman’s Citation Award of Merit to the NextGen Institute’s Equip 2020 Initiative for ADS-B.  Dr. Ligler has also participated in nine National Academies of Sciences, Engineering, and Medicine committees advising the Departments of Transportation, Treasury, and Commerce, and is a current member of the Academies’ Aeronautics and Space Engineering Board. He holds a doctorate in mathematics from the University of Oxford, with his studies supported by a Rhodes Scholarship.

Recent TAMU article on Dr. Ligler: HERE

More on CESG Seminars: HERE

Please join on Friday, 10/7/22 at 10:20 a.m. in ETB 1020.

 

Friday, October 21, 2022
10:20 – 11:10 a.m. (CST)
Virtual via Zoom: https://tamu.zoom.us/j/93347193479 (password in emails or syllabus)

Dr.  Bhuvana Krishnaswamy
Assistant Professor in Electrical and Computer Engineering
University of Wisconsin-Madison

Title: “Scalability in Low Power Wide Area Networks”

Talking Points:
* Long range and low power requirements have contrasting demands on the network.
* Current LPWAN solutions do not satisfy the large-scale needs of practical deployments.

Abstract
Wireless data delivery over long distance is power consuming and challenging for large-scale deployments. Low-power wide area networks (LPWAN) are increasingly in need to develop wireless solutions that satisfy the following requirements (1) Increased battery life, (2) Longer communication range, (3) Large-Scale, and (4) Low-cost. Existing strategies for addressing low-power and long-range do not efficiently address all of these in a large-scale network. In this talk, the fundamental challenges in meeting the above needs of LPWANs will be identified. Collision resolution approaches to meet the demands of large-scale, commercially available LPWANs – LoRa – will be the focus of this talk.

Biography 
Dr. Bhuvana Krishnaswamy is an Assistant Professor in the Department of Electrical and Computer Engineering at University of Wisconsin-Madison. She obtained her MS and Ph.D. in from Georgia Institute of Technology. She is the recipient of NSF CAREER Award, N2Women Rising Star Award, and the Grainger Faculty Scholarship Award. Her research interests are in low-power wireless communication and its challenges in practical deployments. Additional details about her research and team can be found at https://uwconnect.ece.wisc.edu/

More information on Dr. Krishnaswamy can be found HERE.

More info. on past and future CESG Seminars at CESG Seminars (tamu.edu)

* Friday, 10/21/22 at 10:20 a.m. via Zoom *

Friday, October 14, 2022
10:20 – 11:10 a.m. (CST)
Virtual via Zoom: https://tamu.zoom.us/j/93347193479 (password in emails or syllabus)

Dr.  Aditya Mahajan,
Associate Professor in Electrical and Computer Engineering
McGill University – Montreal, Canada

Title: “Approximate Planning and Learning for Partially Observed Systems ”

Talking Points:
* Propose a theoretical framework–based on the fundamental notion of information state–for approximate planning and learning in partially observed systems.
* Define an approximate information state (AIS) and a corresponding approximate dynamic program (ADP).
* Bound the error in using the policy obtained by the solution of the AIS-based ADP.
* Develop RL algorithms based on these bounds and illustrate that they perform well in high-dimensional partially observed grid-world environments.

Abstract
Reinforcement learning (RL) provides a conceptual framework for designing agents which learn to act optimally in unknown environments. RL has been successfully used in various applications ranging from robotics, industrial automation, finance, healthcare, and natural language processing. The success of RL is based on a solid foundation of combining the theory of exact and approximate Markov decision processes (MDPs) with iterative algorithms that are guaranteed to learn an exact or approximate action-value function and/or an approximately optimal policy. However, for the most part, the research on RL theory is focused on systems with full state observations.

In various applications including robotics, finance, and healthcare, the agent only gets a partial observation of the state of the environment. In this talk, I will describe a new framework for approximate planning and learning for partially observed systems based on the notion of approximate information state. The talk will highlight the strong theoretical foundations of this framework, illustrate how many of the existing approximation results can be viewed as a special case of approximate information state, and provide empirical evidence which suggests that this approach works well in practice.

Joint work with Jayakumar Subramanian, Amit Sinha, Raihan Seraj, and Erfan Seyedsalehi

Biography 
Dr. Aditya Mahajan is Associate Professor of Electrical and Computer Engineering at McGill University, Montreal, Canada. He is affiliated with the McGill Center of Intelligent Machines (CIM), Montreal Institute of Learning Algorithms (Mila), and Group for research in decision analysis (GERAD). He received the B.Tech degree in Electrical Engineering from the Indian Institute of Technology, Kanpur, India in 2003 and the MS and PhD degrees in Electrical Engineering and Computer Science from the University of Michigan, Ann Arbor, USA in 2006 and 2008.

He is the recipient of the 2015 George Axelby Outstanding Paper Award, the 2016 NSERC Discovery Accelerator Award, the 2014 CDC Best Student Paper Award (as supervisor), and the 2016 NecSys Best Student Paper Award (as supervisor). His principal research interests are learning and control of decentralized stochastic system.

More information on Dr. Mahajan HERE.

More info. on past and future CESG Seminars at CESG Seminars (tamu.edu)

* Friday, 10/14/22 at 10:20 a.m. via Zoom *