| 학회 | 한국재료학회 |
| 학술대회 | 2019년 가을 (10/30 ~ 11/01, 삼척 쏠비치 호텔&리조트) |
| 권호 | 25권 2호 |
| 발표분야 | 특별심포지엄4. 신개념 광/전자소재의 응용 심포지엄-오거나이저:김연상(서울대) |
| 제목 | Photonic and Probabilistic Artificial Synapses for Energy-efficient Neuromorphic Application |
| 초록 | 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. As a first part, I will present a vertical type two-terminal OHP-based photonic synapse that can mimic the dopamine facilitated synaptic activity. Notably, light illumination on the OHP synaptic device can reduce the onset threshold of the synaptic plasticity, allowing learning and recognition acceleration at a lower power. The accelerated migration of an iodine vacancy inherently existing in the OHP film under the light illumination is considered to be responsible for the further modification of the synaptic behavior. As a second part, I will introduce a vertical form of a gate-tunable probabilistic Si synapse using a SiOx memristor along with a graphene barristor, which was inspired by a tunable and probabilistic synaptic signal processing of the rod-to-rod bipolar cells in the human visual system. Notably, the electrostatic gating from the barristor can modulate the Schottky barrier at the Si/graphene interface; thus, the switching-transition probability and the threshold for signal firing are actively tuned. The device architecture can function as a changeable probabilistic artificial synapse that can be utilized as the basic component of the sparse neural network, which is capable of low-power and learning acceleration for recognizing images. References [1] Ham S. et al. 2019. Adv. Funct. Mater., 29, 1806646 [2] Choi S. et al. 2019. unpublished |
| 저자 | 왕건욱 |
| 소속 | 고려대 |
| 키워드 | Artificial synapse; Neuromorphic computing; memristor |
