Seminar & Colloquium
[세미나: 7월 24일(수), 오후 4시] Prof. Gunuk Wang, Korea University
Title
Brain-inspired Electronic Devices for Artificial Intelligence
Speaker
Prof. Gunuk Wang, KU-KIST Graduate School of Converging Science & Technology, Department of Integrative Energy Engineering, Korea University
* Education
- 2012 Ph.D. in Materials Science and Engineering, Gwangju Institute of Science and Technology (GIST), Korea
- 2007 M.S. in Materials Science and Engineering, Gwangju Institute of Science and Technology (GIST), Korea
- 2005 B.S. in Physics, SungKyunKwan University (SKKU), Korea
* Professional Experience
- Feb. 2023 - Present Department Head, Department of Integrative Energy Engineering, Korea University
- Mar. 2015 - Present Professor, Associate Prof., Assistant Prof., KU-KIST Graduate School of Conversing Science & Technology, Korea University
- Mar. 2020 - Present Professor, Associate Prof., Department of Integrative Energy Engineering, Korea University
- Mar. 2021 - Aug. 2021 Visiting Professor, Korea Institute of Science and Technology
- Sep. 2021- Feb. 2022 Visiting Professor, Seoul National University
- 2012 - 2015 Postdoctoral Research Associate, Department of Chemistry and the Smalley Institute for Nanoscale Science and Technology RICE University, USA (Supervisor: Prof. James M. Tour)
| Date | Wednesday, July 24th , 2024
| Time | 16:00 ~
| Venue | 33동 125호(WCU 다목적실)
[Abstract]
For sustainable advancements in electronics technology, the field of neuromorphic electronics,
i.e., electronics that imitate the principle behind biological synapses with a high degree of
parallelism, has recently emerged as a promising candidate for novel computing technologies.
Toward realizing a massively parallel neuromorphic system, it is essentially required to develop
an artificial synapse capable of emulating various synaptic functionality, such as short- and
long-term synaptic plasticity with ultralow power consumption and robust controllability. In
this talk, as a first part, I will review the general introduction of neuromorphic hardware
technology based on the research background and brain-inspired synaptic device requirements
for high-performance and low-power artificial neural networks, followed by recent results in
this field. As a second part, I will briefly introduce our recent approaches and achievements for
artificial synapses/neurons, diagonal neural network architectures (DCNN), and
probabilistic/reservoir computing using diverse functional nanomaterials (metal (or Si)-oxide
and ferroelectric materials) on advanced device architectures [1-5]. Finally, I will present our
recent study for finger-writing motion recognition in a three-dimensional free-space [6].
References
[1] S. Choi et al., Adv. Mater. 20044659 (2020), Adv. Mater. 34, 2104598 (2022).
[2] S. Choi et al., Nano Energy, 84, 105947 (2021), Adv. Sci, 2104773 (2022).
[3] J. Jang et al., Adv. Sci. 2201117 (2022).
[4] S. Ham et al., Science Adv. 6 : eaba1178 (2020), Nano Energy, 124, 109435 (2024).
[5] S. Choi et al., Nature Comm. 15, 2044 (2024).
[6] H. Cho et al., Nature Electronics. 6, 619-629 (2023).
| Host | 손준우 교수(02-880-7195)