Neuromorphic Computing is a hardware-software co-design approach to realizing Artificial Intelligence (AI) by emulating brains.
Human brains can process real-time signals with remarkably low power consumption (about 20 Watts). Brains consist of a large number of neurons and synapses that operate in parallel (different regions) and communicate with each other through discrete spikes, leading to extraordinarily high computational efficiency. These spiking signals are referred to as Membrane Potential. The firing frequency of the membrane potentials in the nervous system is as low as ~ kilohertz level with millivolt-level magnitudes. Thus, a neuromorphic system can be ultra-high energy-efficient for large-scale artificial neural networks. Consequently, the neuromorphic system is capable of energy-restrained applications, such as edge computing, wearable devices, autonomous vehicles, unmanned aerial vehicles, spacecraft, and so on.
Additionally, biological neural systems have a self-learning capability referred to as associative memory and learning, allowing animals to deal with dynamic environments and plan optimal responses. Unlike prevailing deep learning relying on massive and labeled data, associative learning correlates concurrent events and memorizes the relationship between them. Consequently, the neuromorphic system with associative learning may eventually lead to self-learning intelligence.
Thus, our group aims for several specific topics:
Neuromorphic Robots: Energy-Efficient and Self-Learning Robots through Neuromorphic Systems and Associative Learning/Memory.
Objective: Our group aims to leverage neuromorphic intelligence in robots to achieve a self-learning, real-time, and energy-efficient neuromorphic robot system. Specifically, the sensory signals of robotics, such as Lidar and cameras, will be encoded into spiking signals for routine tasks such as perception, learning, decision-making, and navigation. The embodied neuormorphic AI are expected to have reasoning and common sense via associative learning and memory.
Relevant Publication:
Noah Zins, Yan Zhang, Hongyu An, “Implementation of Associative Memory Learning in Mobile Robots using Neuromorphic Computing” in Neuromorphic Computing, IntechOpen, 2023.
Noah Zins and H. An, "Reproducing Fear Conditioning of Rats with Unmanned Ground Vehicles and Neuromorphic Systems," 2023 24th International Symposium on Quality Electronic Design (ISQED), San Francisco, CA, USA, 2023, pp. 1-7, doi: 10.1109/ISQED57927.2023.10129317.
H. An, Qiyuan An, Y. Yi, “Realizing Behavior Level Associative Memory Learning through Three-dimensional Memristor-based Neuromorphic Circuits”, IEEE Transactions on Emerging Topics in Computational Intelligence, 2019.
2. Neuromorphic Neuroprosthetics
Objective: our group is developing high-power efficient and intelligent neural prosthetics using neuromorphic devices.
Relevant Publication:
Md Abu Bakr Siddique, Yan Zhang, and Hongyu An. "Monitoring Time Domain Characteristics of Parkinson's Disease using 3D Memristive Neuromorphic System." Frontiers in Computational Neuroscience, Volume 17, 2023. DOI: 10.3389/fncom.2023.1274575.
Kerman, Zachary, Chunxiu Yu, and Hongyu An. "Beta Oscillation Detector Design for Closed-Loop Deep Brain Stimulation of Parkinson’s Disease with Memristive Spiking Neural Networks.", 23rd International Symposium on Quality Electronic Design (ISQED), pp. 1-6. IEEE, 2022.
3. Energy-efficient Neuromorphic Electronic Circuit Design for Artificial Neural Networks and Deep learning
Objective: our group is developing novel energy-efficient neuromorphic electronic circuits responding to the extraordinary computational demands for neural network-based artificial intelligent systems. These neuromorphic electronic circuits combine memristive synapses, CMOS-based neurons, and Spiking Neural Networks.
Relevant Publication:
Fabiha Nowshin, Hongyu An, Yang Yi, “Towards Energy-Efficient Memristor-based Spiking Neural Network Accelerators”, ACM Journal on Emerging Technologies in Computing Systems, 2023. in publishment.
H. An, M. S. Al-Mamun, M. K. Orlowski, L. Liu, and Y. Yi, “Three-dimensional Neuromorphic Computing System with Two-layer and Low-variation Memristive Synapses,” in IEEE Transactions on Computer-Aided Design of Integrated Circuits and Systems (TCAD), DOI: 10.1109/TCAD.2021.3061481.
H. An, Mohammad Shah Al-Mamun, Marius K. Orlowski, Lingjia Liu, Yang Yi, “Robust Deep Reservoir Computing through Reliable Memristor with Improved Heat Dissipation Capability,” IEEE Transactions on Computing Aided Design of Integrated Circuits and Systems (TCAD), 2020.
Acknowledgement
