Join us at our next event.

Topic: Learning with Graphs

Next in our spring seminar series, we welcome Dr. Gonzalo Mateos, an Associate Professor from the University of Rochester.

Abstract:
This talk is broadly about learning from network data, which arises for instance with applications involving online social media, recommendation systems, transportation, and network neuroscience. By fruitfully exploiting the inductive biases in relational data, graph neural networks (GNNs) have attained unprecedented performance in various machine learning tasks, including node/graph classification, link prediction, and graph generation. To provide additional motivation, I will start with a user-friendly and didactic introduction to graph signal processing. The goal is to establish the foundations and basic concepts that will be useful to introduce graph GNNs in an intuitive way. After discussing architectures and key properties that make GNNs the model of choice when it comes to learning from relational data, I will highlight several success stories of GNN-based learning for Amazon’s recommendation system, Google Maps navigation, antibiotic discovery, and our own work on explainable brain age prediction.

You can see the presentation live at the UTSA Main Campus in the Student Union Mesquite room (2.01.24).

Those who cannot attend in-person are welcome to tune in virtually through Zoom: https://utsa.zoom.us/j/94807623288.

Topic: Toward Secure, Resilient, and AI-Driven Machine-to-Machine Wireless Communications

Next in our spring seminar series, we welcome Dr. George Sklivanitis, a Research Associate Professor from the Florida Atlantic University.

Abstract:
Next-generation (NextG) network systems such as 6G cellular, future versions of WiFi, satellite networks are expected to connect billions of heterogeneous Internet of Thing (IoT) devices along with billions of people, enable machine-to-machine communications, and provide low-latency computational and storage resources on-demand at the edge and in the cloud. Any failure, tampering or degradation in network service can have highly disruptive, if not potentially catastrophic, effects. NextG network systems should be secure, intelligent and support autonomous decision-making. In this talk, I will present our work on model-based algorithms for waveform design for NextG wireless systems that autonomously optimize how to exchange information with any desirable level of interference avoidance allowed by the physics of the medium and hardware limitations. I will explore challenges and opportunities associated with AI-driven spectrum sensing architectures that provide the support for NextG waveform design and optimization. I will also cover data-driven frameworks enabling NextG security at the physical layer with functionalities such as radio fingerprinting for device authentication and dynamic link encryption using wireless channel state information. I will present our implementation work and experiments with software-defined radio platforms and discuss the way forward.

You can see the presentation live at the UTSA Main Campus in the Student Union Mesquite room (2.01.24).

Those who cannot attend in-person are welcome to tune in virtually through Zoom: https://utsa.zoom.us/j/94807623288.

Topic: Artificial Intelligence Architectures for Grid Data Analysis and Safe Control of Electric Power Systems

Next in our spring seminar series, we welcome Dr. Anna Scaglione, a Professor and IEEE Fellow in the Department of Electrical and Computer Engineering at Cornell Tech.

Abstract:
There is a flurry of research activities recently focusing on applications of Artificial Intelligence in electric power systems analysis and operation. It is often motivated by emerging trends that are tied to decarbonizing the grid through the adoption of distributed renewable energy sources and demand response programs that call for methods account for the stochasticity of the net-demand. The vast majority of the research takes AI tools as a black box that internalizes every aspect of the observations, including the physics of the system (namely Ohm’s law) and learns to control the system by being rewarded in choosing good actions. In this talk we discuss how to go a step further. First we introduce Complex Graph Neural Network and show that they are most effective in both inference and reinforcement learning tasks, thanks to their physics inspired feature extraction capabilities. Then we move to introduce the problem of “safe” reinforcement learning, which allows to enforce the physical and safety constraints in the optimal policy.

You can see the presentation live at the UTSA Main Campus in the Student Union Mesquite room (2.01.24).

Those who cannot attend in-person are welcome to tune in virtually through Zoom: https://utsa.zoom.us/j/94807623288.

Topic: Sparse Sensor Array Processing for High-Resolution Sensing

Next in our spring seminar series, we welcome Dr. Yimin D. Zhang, an Associate Professor in the Department of Electrical and Computer Engineering at Temple University.

Abstract:
Array signal processing plays a crucial role in various applications, including radar sensing, wireless communications, and medical imaging. It serves as a vital technology for tasks like directional beamforming, interference suppression, multiuser access, direction finding, signal localization, and computational imaging. The evolving needs of modern radar sensing and wireless communications demand higher spatial resolution and improved estimation accuracy. However, these requirements often translate into a significant hardware cost due to the large array aperture and the high number of array sensors.To address these challenges, sparse sensor array design and processing offer low-complexity solutions that enhance sensing capability and improve performance with a reduced number of array sensors. This presentation explores recent advancements in concepts, design strategies, and processing methods for sparse array-based sensing, aiming to achieve high spatial resolution and an increased number of degrees of freedom. These advancements leverage recent progress in compressive sensing, optimization, and machine learning. Various sparse array processing techniques will be discussed and array designs that reduce difference lag redundancies and exploit frequency diversity will be presented, providing insights into efficient and cost-effective solutions for array signal processing in diverse applications.

You can see the presentation live at the UTSA Main Campus in the Student Union Mesquite room (2.01.24).

Those who cannot attend in-person are welcome to tune in virtually through Zoom: https://utsa.zoom.us/j/94807623288.

Topic: Contractive Diffusion Models and Score Matching by Continuous Reinforcement Learning

Next in our spring seminar series, we welcome Dr. Wenpin Tang, an Assistant Professor in the Department of Industrial Engineering and Operations Research at Columbia University.

Abstract:
In this talk, I will link two different topics. The past decade has witnessed the success of generative modeling (e.g., GANs and VAEs) in creating high quality samples in a wide variety of data modalities. The first part of this talk is concerned with the recently developed diffusion models, the key idea of which is to reverse a certain stochastic dynamics. I will first take a continuous-time perspective and examine the performance of different SDE schemes including VE (variance exploding) and VP (variance preserving). The discretization is more subtle, and our idea is to “contract” the reversed dynamics leading to possible new diffusion model designs. In the second part, I will talk about continuous reinforcement learning. Reinforcement Learning (RL) has been successfully applied to wide-ranging domains in the past decade. In recent years, a fast-growing body of research has extended the frontiers of continuous RL to designing model-free methods and algorithms. I will discuss the recently introduced “q-learning” and closely related policy optimization. Finally, I will highlight a natural application of continuous RL to fine-tune the score function in the diffusion models.

You can see the presentation live at the UTSA Main Campus in the Student Union Mesquite room (2.01.24).

Those who cannot attend in-person are welcome to tune in virtually through Zoom: https://utsa.zoom.us/j/94807623288.

Topic: Data Science for Translational Research

Next in our fall seminar series, we welcome Dr. Debashis Sahoo, an Associate Professor in the Pediatrics and Computer Science and Engineering at the University of California San Diego.

Abstract:
Translational research aims to swiftly apply basic science discoveries in practical healthcare. Productivity hurdles can impede the quick translation of research into healthcare breakthroughs, with issues like reproducibility and generalizability of biological results in clinical settings posing significant challenges. Concerns about irreproducible results are on the rise, with over 70% of researchers reportedly facing challenges reproducing others’ experiments, and more than half struggling with their own. Drug development, a crucial facet of translational research, maintains a low success rate despite past efforts.
Part of the problem lies in the tools of data science and statistics used. I’ll introduce you to a new approach—Boolean analysis of large and diverse biological datasets. In my research group, we’ve pioneered this method and developed computational tools to uncover invariant logical relationships between genes. Join me as we explore how leveraging these invariant relationships can enhance the reproducibility of current approaches. We will discuss how Boolean analysis can transform the power of AI/ML approaches.

You can see the presentation live at the UTSA Main Campus in the Student Union Hidalgo room (2.214).

Those who cannot attend in-person are welcome to tune in virtually through Zoom: https://utsa.zoom.us/j/91049981814.