CS Seminar by Prof. Yoon Seok Yang (Friday, October 31, 2 PM, B204)
Writer Computer ScienceDate Created 2025.10.24Hits15
Prof. Yoon Seok Yang will be giving a talk on "MindCore: Spike-Driven Programmable Accelerator for On-device Neuromorphic Computing".
Please find the seminar details below: Title: MindCore: Spike-Driven Programmable Accelerator for On-device Neuromorphic Computing Date & Time: Friday, October 31, 2 PM Venue: B204
Abstract
Spiking neural networks (SNNs) are drawing attention as a promising way to achieve low-power, real-time intelligence on edge devices. In this presentation, we introduce MindCore, a spike-driven accelerator designed to support on-device neuromorphic computing. MindCore builds on a streamlined Reduced Instruction Set Compute (RISC) processor and provides programmable neurons and synapses, allowing flexibility across a wide range of SNN models. To handle spike-based operations efficiently, we developed a set of custom instructions, including Single Instruction Multiple Data (SIMD) accumulators for 8-, 16-, and 32-bit operations, and a spike-vector instruction that performs logical AND, spike counting, and accumulation in one step. The proposed design was implemented on a Xilinx ZCU104 FPGA and verified through Register Transfer Level (RTL) testing, together with a software toolchain for model compilation and application support. Demonstrations on vision inference and dementia diagnosis assistance confirm that MindCore achieves meaningful energy savings without sacrificing accuracy, showing its potential as a practical platform for future on-device neuromorphic AI.
Speaker Bio
Prof. Yoon-Seok Yang is an assistant professor in the Computer Science department. He previously worked as a Tensor Processing Unit (TPU) silicon and research engineer at Google in Sunnyvale, California, USA. Before joining Google, he served as a research scientist at the Neuromorphic Computing Lab at Intel Labs in Santa Clara, California, USA from 2012 to 2022, with a research focus on neuromorphic computing systems and AI chip design. Prof. Yang holds a Ph.D. in electrical and computer engineering from Texas A&M University, College Station, Texas, USA.