Neuromorphic Computing and Brain-Computer Interface System

Members: Yeongwoo Jang, Daye Jung, Seunghyun Song, Haewon Kim

Motivation

The brain is an extraordinarily complex system, and understanding its mechanisms is essential for advancing neuroscience and enabling transformative real-world applications. To address this challenge, we pursue two complementary research directions.

Neuromorphic Computing: Despite remarkable progress in AI inspired by the brain, we still lack the understanding of how a biological brain actually operates. This gap limits both scientific discovery and future advances in brain-inspired computing. To address this, we are building a special-purpose computer system to faithfully simulate and analyze a human-scale complex brain. We expect our system to serve as a crucial tool for uncovering the principles of brain function and advancing next-generation AI.
Brain-Computer Interface (BCI) System: At the same time, understanding the brain also requires observing and interacting with it in real environments. Scalable and efficient BCI systems are essential for studying neural activity and supporting clinical applications (e.g., seizure detection, tremor suppression, visual prosthesis). To meet these needs, we are developing high-performance BCI systems that can operate reliably and efficiently in practical settings.

Research

Brain Simulation System [ISCA’18, MICRO’19, ASPLOS’21, HPCA’22]. Brain simulation is a key process to understand the detailed working mechanism of a complex brain. A human-scale brain consists of billions of heterogeneous neurons interconnected via synapses, and its communication is based on complex spike deliveries. Therefore, conventional computers are too slow or expensive to faithfully simulate a brain, whereas custom-design brain simulation systems suffer from their limited model coverage. To resolve the issue, we are building a fast and scalable brain-simulation system. In this regard, we propose Flexon and FlexLearn, two cost-effective ASIC-based architectures to dynamically construct heterogeneous biological neuron models and learning rules, respectively. We are also developing FlexBrain, a fast, accurate, and scalable full-stack brain simulation system that efficiently incorporates Flexon and FlexLearn ASICs. We plan to release the RTL code for Flexon and FlexLearn datapath as well. Moreover, we optimize the single-core architecture by adopting an event-driven simulation method. The resulting design, NeuroEngine, significantly improves the simulation speed and energy efficiency.

Applying to AI/Deep Learning [Neurocomputing’21]. A human brain is extremely power efficient when compared to modern AI accelerators. Therefore, the brain’s spiking neural network (SNN) mechanism has been considered highly promising to improve the power efficiency of modern AI accelerators. But, such accurate and power-efficient SNN-based AI accelerators do not exist yet because we do not know how to maintain their advantages while accurately processing AI/DL applications. To resolve the issue, we are developing cost-effective SNN-based AI/DL execution mechanisms and acceleration systems.

Brain-Computer Interface (BCI) Systems [MICRO’24, ISCA’25, ASPLOS’26]. BCI systems enable direct communication between the brain and external devices by capturing and processing neural signals. As BCIs evolve to support more electrodes and complex tasks, they encounter new challenges—including the need for flexible, energy-efficient signal processing, as well as robust adaptation to non-stationary and dynamic neural patterns over time. To address these issues, we propose two platforms: NeuroLobe and InfiniMind, which target complementary bottlenecks in next-generation BCIs. NeuroLobe is a flexible, event-driven neuromorphic processor designed for spike-driven BCI workloads, which enables efficient and scalable support on diverse BCI algorithms. Also, InfiniMind focuses on continual learning to handle non-stationarity in neural signals, with optimizations that reduce non-volatile memory writes to extend device lifetime and boost learning performance. Building on these advances, we are also developing new architectural and system-level innovations to improve scalability and usability in real-world BCI applications. To further these innovations, we have released TierX, the first simulation framework for multi-tier BCI system design. Multi-tier BCI processing leverages the unique strengths of various computing tiers, each with distinct constraints on power budget and communication efficiency. TierX empowers BCI system architects to navigate the complex design space and derive the optimal multi-tier system for any BCI workload.

Software release

Publications

  • TierX: A Simulation Framework for Multi-tier BCI System Design Evaluation and Exploration
    Seunghyun Song, Yeongwoo Jang, Daye Jung, Kyungsoo Park, Donghan Kim, Gwangjin Kim, Hunjun Lee, Jerald Yoo, and Jangwoo Kim
    ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), Mar. 2026
  • InfiniMind: A Learning-Optimized Large-Scale Brain-Computer Interface
    Yeongwoo Jang*, Daye Jung*, Seunghyun Song, Hunjun Lee, and Jangwoo Kim
    ACM/IEEE International Symposium on Computer Architecture (ISCA), Jun. 2025
  • Rearchitecting a Neuromorphic Processor for Spike-Driven Brain-Computer Interfacing
    Hunjun Lee, Yeongwoo Jang, Daye Jung, Seunghyun Song, and Jangwoo Kim
    ACM/IEEE International Symposium on Microarchitecture (MICRO), Nov. 2024
  • NeuroSync: A Scalable and Accurate Brain Simulation System using Safe and Efficient Speculation
    Hunjun Lee*, Chanmyeong Kim*, Minseop Kim, Yujin Chung, and Jangwoo Kim
    25th IEEE International Symposium on High-Performance Computer Architecture (HPCA), Apr. 2022
  • An Accurate and Fair Evaluation Methodology for SNN-Based Inferencing with Full-Stack Hardware Design Space Explorations
    Hunjun Lee, Chanmyeong Kim, Eunjin Baek, and Jangwoo Kim
    Neurocomputing, Sep 2021
  • NeuroEngine: A Hardware-based Event-driven Simulation System for Advanced Brain-inspired Computing
    Hunjun Lee*, Chanmyeong Kim*, Yujin Chung, and Jangwoo Kim
    ACM International Conference on Architectural Support for Programming Languages and Operating Systems (ASPLOS), Apr. 2021
  • FlexLearn: Fast and Highly Efficient Brain Simulations Using Flexible On-Chip Learning
    Eunjin Baek*, Hunjun Lee*, Youngsok Kim, and Jangwoo Kim
    ACM/IEEE International Symposium on Microarchitecture (MICRO), Oct. 2019
  • Flexon: a flexible digital neuron for efficient spiking neural network simulations
    Dayeol Lee*, Gwangmu Lee*, Dongup Kwon, Sunghwa Lee, Youngsok Kim, and Jangwoo Kim
    ACM/IEEE International Symposium on Computer Architecture (ISCA), June. 2018

* Contributed equally