Kim Jin Hong, Kim Hyun Wook, Chung Min Jung, Shin Dong Hoon, Kim Yeong Rok, Kim Jaehyun, Jang Yoon Ho, Cheong Sun Woo, Lee Soo Hyung, Han Janguk, Park Hyung Jun, Han Joon-Kyu, Hwang Cheol Seong
Department of Materials Science and Engineering and Inter-University Semiconductor Research Center, Seoul National University, Gwanak-ro 1, Gwanak-gu, Seoul 08826, Republic of Korea.
System Semiconductor Engineering and Department of Electronic Engineering, Sogang University, 35 Baekbeom-ro, Mapo-gu, Seoul 04107, Republic of Korea.
Nanoscale Horiz. 2024 Nov 19;9(12):2248-2258. doi: 10.1039/d4nh00421c.
In-sensor computing has gained attention as a solution to overcome the von Neumann computing bottlenecks inherent in conventional sensory systems. This attention is due to the ability of sensor elements to directly extract meaningful information from external signals, thereby simplifying complex data. The advantage of in-sensor computing can be maximized with the sampling principle of a restricted Boltzmann machine (RBM) to extract significant features. In this study, a stochastic photo-responsive neuron is developed using a TiN/In-Ga-Zn-O/TiN optoelectronic memristor and an Ag/HfO/Pt threshold-switching memristor, which can be configured as an input neuron in an in-sensor RBM. It demonstrates a sigmoidal switching probability depending on light intensity. The stochastic properties allow for the simultaneous exploration of various neuron states within the network, making identifying optimal features in complex images easier. Based on semi-empirical simulations, high recognition accuracies of 90.9% and 95.5% are achieved using handwritten digit and face image datasets, respectively. In addition, the in-sensor RBM effectively reconstructs abnormal face images, indicating that integrating in-sensor computing with probabilistic neural networks can lead to reliable and efficient image recognition under unpredictable real-world conditions.
作为克服传统传感系统中固有的冯·诺依曼计算瓶颈的一种解决方案，传感器内计算已受到关注。这种关注源于传感器元件能够直接从外部信号中提取有意义的信息，从而简化复杂数据。利用受限玻尔兹曼机（RBM）的采样原理来提取显著特征，可使传感器内计算的优势最大化。在本研究中，使用TiN/In-Ga-Zn-O/TiN光电忆阻器和Ag/HfO/Pt阈值开关忆阻器开发了一种随机光响应神经元，其可配置为传感器内RBM中的输入神经元。它表现出取决于光强度的S形开关概率。随机特性允许在网络内同时探索各种神经元状态，从而使在复杂图像中识别最佳特征变得更容易。基于半经验模拟，使用手写数字和面部图像数据集分别实现了90.9%和95.5%的高识别准确率。此外，传感器内RBM有效地重建了异常面部图像，这表明将传感器内计算与概率神经网络相结合可在不可预测的现实世界条件下实现可靠且高效的图像识别。
