13th November, 2024 from 3:00PM – 4:00PM (GST)
| Title: | Pushing Boundaries in Spiking Neural Networks |
| Location: | Wave Auditorium |
| Speaker: | SNN research group at MBZUAI |
| Abstract: | The SNN research group at MBZUAI, established three years ago, focuses on advancing both the theoretical and practical aspects of Spiking Neural Networks (SNNs) and aims to build a comprehensive local research center for neuromorphic computing. Since its inception, the group has published 10 research papers in top-tier conferences (NeurIPS, ICML, ICLR, AISTATS, AAAI) and prestigious journals (JMLR). Members of the group actively collaborate with renowned research centers worldwide. The seminar will begin with a brief overview of SNNs, exploring their foundational concepts and the potential benefits of their implementations. Following this, we will present key research directions pursued within our group, underscoring our contributions to advancing the field. To conclude, we will showcase a live demo featuring an SNN model implemented on neuromorphic hardware, trained to recognize gestures through a DVS (Dynamic Vision Sensor) camera. |
Presentors
Dr. Bin Gu Bin Gu is currently an assistant professor in the Department of Machine Learning at the Mohamed bin Zayed University of Artificial Intelligence (MBZUAI). He was previously a Postdoc at Pittsburgh University, Taxes University at Arling, and Western University. He also held a professor position at Nanjing University of Information Science and Technology. His research interests include large-scale optimization in machine learning and spiking neural networks.  |
Dr. Velibor Bojkovic Velibor Bojkovic is currently a research fellow at Machine Learning Department at MBZUAI. He holds a PhD in pure mathematics from University of Padova (Italy) and University of Bordeaux (France), and has previously worked as a researcher at University of Padova, Haifa University in Jerusalem, University of Caen Normandy. His research interest include arithmetic geometry, spiking neural networks, functioning, interpretability and reasoning in llms.  |
Dr. Bhaskar Mukhoty Bhaskar Mukhoty is currently a postdoctoral associate at MBZUAI. He obtained his PhD in Machine Learning from the Department of CSE, IIT Kanpur. His research interests include developing scalable machine learning / deep learning algorithms with theoretical guarantees on possibly corrupted training and test data, with a broad interest in understanding competing aspects of prediction and security, and spiking neural networks, where he studied the implications of this energy-efficient neuronal computational model on generalization and adversarial robustness tasks.  |
Dr. Hilal AlQuabeh Hilal AlQuabeh is a postdoctoral researcher in Natural Language Processing at MBZUAI, specializing in machine learning optimization and interpretability. With a PhD in Machine Learning from MBZUAI and an MSc in Mechanical Engineering from MIT-Masdar Institute, Hilal's work spans spiking neural networks, state-space models, neural network optimization, and advancing AI robustness.  |
Yasser Ashraf Yasser Ashraf is a master's student in the Machine Learning Department at MBZUAI with a bachelor's in Mechatronics and Robotics Engineering from Egypt-Japan University of Science and Technology. Yasser’s research interests span Robotics, Spiking Video understanding, and real-time Spiking Neural Networks applications.  |