Spring & Fall 2026 | Mondays | Room 5150 | Zoom Link
| Date | Speaker | Research Area / Topic | Room |
|---|---|---|---|
| Apr 6, 2026 | Dr. Haihan Nan | Neural Radiance Fields for Wireless Channel Modeling | 5150 |
| Apr 13, 2026 | Dr. Seonghoon Park | Real-time VFM/ViT inference for mobile ARAbstract: Mobile augmented reality (AR) applications require real-time visual understanding, including pixel-level depth estimation and semantic segmentation, to enable immersive and context-aware experiences. Vision Transformer (ViT)-based Vision Foundation Models (VFMs) offer strong generalization across diverse environments; however, their deployment on mobile devices remains challenging due to limited computational resources. In this seminar, Dr. Park will present two recent works on efficient VFM inference for mobile platforms, along with ongoing research directions. The first study investigates leveraging heterogeneous mobile processors—including CPUs, GPUs, and NPUs—to optimize VFM inference (ACM MobiSys 2025). The second focuses on NPU-specific optimizations for efficient VFM execution on mobile devices (ACM EuroSys 2026). The seminar will also briefly highlight ongoing research on edge-assisted VFM inference. | 5150 |
| Apr 16, 2026 | Dr. Beibei WangBio: Beibei Wang received the B.S. degree in electrical engineering from the University of Science and Technology of China in July 2004 (with the highest honor), and the M.S. and Ph.D. degrees in electrical engineering from the University of Maryland, College Park in 2008 and 2009, respectively. In 2009-2010 she was a postdoctoral research associate with the University of Maryland, College Park. In 2010-2012, she was with Qualcomm Research and Development, San Diego, working on system design and 3GPP RAN2 aspects of HSPA heterogeneous networks. In 2012-2014, she was with Qualcomm Research Center at New Jersey, working on system design and 3GPP RAN1 aspects of LTE-Direct. Since 2015, she has been with Origin AI, where she is Vice President Research. Her research interests include wireless sensing and positioning, Internet of Things, and mobile computing. She is a co- author of “Wireless AI: Wireless Sensing, Positioning, IoT, and Communications,” Cambridge University Press, 2019 and “Cognitive Radio Networking and Security: A Game Theoretic View,” Cambridge University Press, 2011. Her experience includes serving as an IEEE SPS Distinguished Industry Speaker (2026-2027), a Sensors Council Distinguished Lecturer (2025-2027), and a recipient of the IEEE Sensors Council Technical Achievement Award in Sensor Systems or Networks - Advanced Career (2025). She is a fellow of IEEE. | Wi-Fi Can Do More: Towards Ubiquitous Wireless SensingAbstract: The next big deal for Wi-Fi is not about communication and networking, but sensing. Wireless Sensing technology is turning a Wi-Fi device into a ubiquitous sensor, which not only adds a brand-new dimension to the functions, capabilities, and applications of all Wi-Fi systems, but also revolutionizes how sensing, especially human-centric sensing, is practiced. Wi-Fi Sensing utilizes ambient Wi-Fi signals to analyze and interpret human and object movements, underpinning many sensing applications such as motion sensing, sleep monitoring, fall detection, etc. These new sensing functionalities can benefit the global Wi-Fi ecosystem including integrated circuit manufacturers, device manufacturers, system integrators, application developers, and ultimately end users. In this talk, we introduce the concepts, principles of Wi-Fi Sensing, and share our unique technologies that have been deployed for real-world applications. We foresee that Wi-Fi Sensing will enter billions of devices and millions of homes, creating a smarter space for a smarter life. | 5150 |
| Apr 20, 2026 | Md Rubel SarkarBio: Md Rubel Sarkar received the B.Sc. degree in Electrical and Electronic Engineering from Ahsanullah University of Science and Technology, Dhaka, Bangladesh, and the M.Sc. degree in Electrical and Computer Engineering from Virginia Tech, USA. He is currently pursuing the Ph.D. degree in Electrical and Computer Engineering at Virginia Tech. His research interests include in-memory computing, hyperdimensional computing, and reservoir computing. Prior to joining Virginia Tech, he worked for over five years (2017-2022) at the GlobalFoundries Offshore Design Center (ODC) in Bangladesh, where he gained extensive experience in standard cell library development, mixed-signal and memory layout design, and physical design. | NMPHDC: A 12 nm Reconfigurable Multi-port SRAM based Near-Memory Hyperdimensional Computing Architecture.Abstract: Hyperdimensional Computing (HDC) is a brain- inspired computing paradigm that represents information using high-dimensional vectors, referred to as hypervectors (HVs), enabling inherent robustness to noise and efficient learning on resource-constrained platforms. In this work, we present NMPHDC, a reconfigurable multi-port (MP) SRAM-based logic- near-memory (LnM) HDC architecture that supports on-chip training and inference. The proposed design is configurable across multiple HV dimensions, including 256, 512, and 1024 bits, enabling flexibility across diverse workloads. A lightweight custom LnM block enables LnM execution of key HDC operations, including XOR for binding and majority (MAJ) for bundling, significantly reducing data movement overhead. In addition, we propose a custom on-chip training algorithm and integrate a dedicated on-chip random number generator (RNG) to strengthen training security through locally generated HVs. A centralized HDC controller orchestrates dataflow, computation, training, and testing with seamless configurability. Implemented in GlobalFoundries (GF) 12 nm low-power (LP) FinFET technology, NMPHDC occupies 1.25 mm2, consumes ∼0.72 mW in standby and ∼19.7 mW at 100 MHz, and achieves classification accuracies of 93.33%, 25.75%, 72.62%, and 75.18% on the ECG5000, CIFAR-10, MNIST, and ISOLET datasets, respectively. | 5150 |
| Apr 27, 2026 | Ramashish GauravBio: Ramashish Gaurav has been conducting graduate research in SNNs and Neuromorphic Computing for close to 5 years. He has experience of developing novel Reservoir-based SNN architectures, designing SNN training algorithms, and building their novel applications with neuromorphic chips. He strongly believes that SNNs are at the frontiers of next-generation AI systems, especially in the domain of Edge-AI. | The BRain Inspired Computing in BRICCS LabAbstract: This talk on Brain-Inspired Computing is mainly introductory in nature, where we will go through the foundations of Spiking Neural Networks (SNNs) and Neuromorphic Computing. The goal is to engage in proactive discussions and discover synergy among SNNs, Communication, and Security. Unlike Artificial Neural Networks (ANNs) that process continuous values, the SNNs process information based on the computational principles of brain, i.e., the SNNs process action potentials (a.k.a. spikes), thus, they offer the computational sparsity of brain. We will see that the SNNs are architecturally isomorphic to the ANNs and are highly energy-efficient (30-100x ) when deployed on specialized neuromorphic hardware, e.g., Intel’s Loihi-1/2 neuromorphic chips. We will also go through a novel kind of reservoir, called Legendre Delay Network (LDN), and investigate its performance for Time Series Classification (TSC) tasks - when it is combined with SNNs. We will see that such Legendre-based SNNs outperform LSTM-Conv integrated models on more than 31% of 102 TSC datasets. We will briefly look into what makes SNNs suitable for applications in medical, wireless communication, and security related areas. We will also go through the capabilities of neuromorphic chips in our BRICCS lab, and what makes them highly energy-efficient and low-latency substrates. Finally, we will go through their programming principles and how to leverage them to build the next generation AI computing systems - straight from the BRICCS lab! | 3100 |
| Apr 30, 2026 | Dr. Kangjun BaiBio: Dr. Kang Jun Bai serves as a Research Electronics Engineer within the Air Force Research Laboratory Information Directorate (AFRL/RI). He earned his Ph.D. in Electrical Engineering from Virginia Tech in 2021, following a Master of Science and a Bachelor of Science from San Francisco State University. Dr. Bai’s specialized research encompasses mixed-signal integrated circuits & systems design, nanoelectronic devices for neuromorphic computing, and the development of edge applications for mobile intelligence platforms. A prolific researcher, he has authored over 50 publications in premier ACM/IEEE venues. His professional standing is distinguished by multiple prestigious honors, notably the 2023 Society of Asian Scientists and Engineers Achievement Award, the 2023 Bill and LaRue Blackwell Best Dissertation Award from Virginia Tech, and the 2021 Qualcomm Innovation Fellowship. | Bridging Research and the Battlefield: Strategic Opportunities and Technical Horizons at AFRL/RIAbstract: The Air Force Research Laboratory Information Directorate (AFRL/RI) serves as the Department of the Air Force’s preeminent research organization for Command, Control, Communications, Computers, Cyber, and Intelligence (C5I) technologies. The Directorate is tasked with the exploration, prototyping, and demonstration of high-impact, affordable, and transformative technologies designed to ensure a decisive technical advantage for the United States. By converting vast streams of data into actionable intelligence, AFRL/RI empowers decision-makers and fortifies national security across air, space, and cyberspace domains. This presentation provides a comprehensive overview of the AFRL/RI mission and its role in national defense. The discussion will delineate core technical competencies, specifically focusing on the development of sophisticated autonomous computational capabilities for size, weight, and power (SWaP)-constrained platforms. Furthermore, the session will highlight the Directorate's specialized facilities and existing outreach initiatives designed to facilitate strategic collaborations with academic institutions. | — |
| May 4, 2026 | Muhammad Farhan AzmineBio: Muhammad Farhan Azmine is a 4th-year PhD student working on neuromorphic hardware-algorithm co-design for energy-efficient, edge-intelligent neural systems. His research focuses on reservoir-based spiking neural networks and LDN-enhanced reservoir computing for time-series processing, with developing hardware-friendly learning algorithms validated in software and mapped to RTL for FPGA implementation. | A DSP-Optimized Ultra-Power-Efficient Reservoir Computing Alternative for Spike-Driven Neuromorphic ECG Classification ProcessorAbstract: This talk presents a DSP-optimized, ultra-power-efficient ECG classification processor system based on a hardware-software co-design approach for edge wearable devices, addressing the limitations of conventional ANN-based solutions in power efficiency and patient-independent accuracy. The system employs spike-driven computation using Spiking Neural Networks (SNNs) and Liquid State Machines (LSMs) with local learning algorithms to enable adaptive, real-time inference. To improve hardware efficiency, the conventional LSM reservoir is replaced with an intelligent hardware mapping implementation of Legendre Delay Network (LDN)-based feature extractor, enabling compact temporal representation with reduced complexity. The FPGA implementation leverages optimized DSP architecture and achieves up to 86% LUT reduction, 64% flip-flop reduction, and 65% lower dynamic power consumption, while improving classification accuracy by at least 10%, with a moderate latency trade-off. | 3100 |
| May 11, 2026 | Ramin Safavinejad | 3100 | |
| May 13, 2026 | Dr. Hongyu AnBio: Hongyu An is an assistant professor in Electrical and Computer Engineering at Michigan Technological University. He obtained his doctoral degree from Virginia Tech, his master's degree from the University of Science and Technology, and his bachelor's degree from Shenyang University of Technology, all in electrical engineering. His research interests include Neuromorphic Computing, AI Hardware, AI for Human Health and Medical Devices, and Neuromorphic neuroprosthetics. | Neuromorphic Circuits and Systems for Neural Prostheses: From Software-Hardware Co-design to Biological ValidationAbstract: Neurological disorders such as Parkinson’s disease and memory impairments affect millions of people worldwide. Implantable neural prostheses, including Deep Brain Stimulation (DBS) and emerging memory prosthetic systems, have demonstrated strong therapeutic potential. However, most current systems operate in an open loop manner and lack the adaptive capability to respond to dynamic fluctuations in clinical symptoms. Closed loop adaptive neuromodulation offers several key advantages: the ability to track symptom fluctuations in real time, improved energy efficiency through symptom dependent stimulation, and the potential to enable more personalized treatments tailored to patient specific disease dynamics. In this talk, Dr. Hongyu An presents a neuromorphic circuits and systems framework for adaptive neuromodulation. Inspired by principles of brain inspired computing, this work develops adaptive neuromorphic neural prosthesis systems based on silicon neurons, memristive synapses, and closed loop neuromodulation architectures. The research follows an integrated engineering pipeline that connects computational neural modeling, software and hardware codesign methodology, biological validation, and rodent experiments. | 5150 |
| May 18, 2026 | Yibin Liang | Souvik Pramanik | 3100 |
| May 25, 2026 | Memorial Day — No Seminar | — | |
| Jun 1, 2026 | Prof. Yang, Yaling | 5150 | |
| Jun 8, 2026 | Chenyue Wang | 5150 | |
| Jun 15, 2026 | Dr. Karim Said | 5150 | |
| Jun 22, 2026 | Alberta Dadeboe | 5150 | |
| Jun 29, 2026 | Daniel Rosen | 5150 | |
| Jul 6, 2026 | Jan Acosta | 5150 | |
| Jul 13, 2026 | Muhammad Zubair | 3100 | |
| Jul 20, 2026 | Shreya Datir | 5150 | |
| Jul 27, 2026 | Honghao Zheng | 5150 | |
| Aug 3, 2026 | Mihir Patel, Gayathri Mahendran | 5150 | |
| Aug 10, 2026 | Usama Saeed | 5150 | |
| Aug 17, 2026 | Chunxiao Lin | 5150 | |
| Aug 24, 2026 | Raymond Zhao | 5150 | |
| Aug 31, 2026 | Jiongyu Dai | 5150 | |
| Sep 7, 2026 | Labor Day — No Seminar | — | |
| Sep 14, 2026 | Shadab Mahboob | — | |
| Sep 21, 2026 | Ummay Sumaya Khan | 5150 | |
| Sep 28, 2026 | Zipeng Lin | 5150 | |
| Oct 5, 2026 | Evan Allen | 5150 | |
| Oct 12, 2026 | 5150 | ||
| Oct 19, 2026 | Suchismita Batabyal | 5150 | |
| Oct 26, 2026 | Ruizhe Li | 5150 | |
| Nov 2, 2026 | Sujata Sinha | Token Based Rate Distortion Theory for Scaling Laws | 5150 |
| Nov 9, 2026 | 5150 | ||
| Nov 16, 2026 | Ibraheem Alturki | 5150 | |
| Nov 23, 2026 | Shirazush Salekin Chowdhury | 5150 | |
| Nov 30, 2026 | Jiaqiang Ling | 5150 | |
| Dec 7, 2026 | 5150 | ||
| Dec 14, 2026 | Zeyuan Hou | 5150 | |
| Dec 21, 2026 | Nima Mohammadi | 5150 | |
| Dec 28, 2026 | Xiaomeng WangBio: Xiaomeng Wang received the B.S. degree in Elec- trical Engineering from Beijing Jiaotong University, Beijing, China, and the M.S. degree in Electrical Engineering from The University of Hong Kong, Hong Kong, China. He is currently pursuing the Ph.D. degree in Computer Engineering with the Bradley Department of Electrical and Computer Engineering, and with the Institute for Advanced Computing, Virginia Tech, Alexandria, VA, USA. His research interests include machine learning for electronic design automation, VLSI design optimization, and hardware acceleration for generative AI training and inference. More information is available at www.wangxm.com. | — | |