Prof. Jin-Hong Park’s Research Team developed artificial intelligence (AI) semiconductor: optoelectronic synaptic device
Electronic and Electrical Engineering
Prof. PARK, JIN HONG
Dr. Seunghwan Seo & Dr. Je-Jun Lee
Advanced Materials 5th October (IF: 12.121); An optogenetics-inspired flexible van-der-Waals optoelectronic synapse and its application to convolutional neural network; Seunghwan Seo (1st author) / Je-Jun Lee (1st author) / Prof. Jin-Hong Park (Corresponding author)
□ Neuromorphic chips that mimic the learning principle of the human brain are in the spotlight as next-generation processing chips because they can minimize power consumption by processing a large amount of information in parallel and improve their computational functions through learning. In particular, various researches on synaptic semiconductor devices, which are essential for realizing parallel information processing and learning capabilities of neuromorphic chips, have been actively conducted worldwide.
□ The research team implemented a flexible optoelectronic synapse on 2D vdW layered rhenium disulfide, which features an inherent photosensitive memory nature derived from the intrinsic persistent photoconductivity effect. Then, the research team successfully demonstrated the applicability of the flexible 2D vdW optoelectronic synapse to optogenetics-inspired intelligent systems via training and inference tasks using a convolutional neural network and the CIFAR-10 dataset.
□ These studies were published in the Advanced Materials on October 5th.