Description
Despite significant advances in computing, we remain distant from replicating the efficiency, speed, and intelligence of natural brains. Current computing architectures are increasingly constrained by fundamental limitations, such as the memory bottleneck and the diminishing returns of performance improvements due to the end of Dennard scaling. These challenges have prompted researchers worldwide to explore brain-inspired computational models. Neuromorphic computing has emerged as a promising approach, offering energy-efficient hardware designed to handle real time signal processing and support a variety of edge AI applications. This method harnesses event-driven, massively parallel spiking neural networks to achieve brain like computation at the hardware level.In my talk, I will explore the fundamentals of spiking neural networks, focusing on modern training techniques and their hardware implementations that prioritize event-driven sensing and computing. Neuromorphic systems enable ultra-low-power embedded implementations, making them ideal for a wide range of smart sensing applications. I will wrap up by showcasing several prototype devices that meet the stringent energy and cost-saving demands of the Internet of Things (IoT), with applications in biomedical signal processing. These prototypes illustrate the potential of spiking neural networks in advancing embedded sensing and computing technologies.
| Period | 28 Aug 2024 |
|---|---|
| Event title | Eindhoven Taiwan Summer School : Advancing innovations in semiconductor electronics & photonics |
| Event type | Workshop |
| Location | Eindhoven, NetherlandsShow on map |
| Degree of Recognition | International |
Keywords
- semiconductor electronics
- neuromorphic engineering