BROWSE
Detail
Journal Article
In Situ Integrated Titanium Oxide Synaptic Phototransistor Enabling Multimodal Plasticity and Noise-Robust Selective Attention
- Authors
- Youngbin Yoon, Jaehee Lee, Jeongwook Kwon, Yongki Kim, Myunghun Shin, Jung Wook Lim
- Citation
- Advanced Materials Technologies, Early Access
- ISSN
- 2365-709X
- Publisher
- John Wiley & Sons
- Language
- English
- Type
- Journal Article
- Abstract
- Multimodal synaptic plasticity with temporal responsiveness is essential for the construction of energy-efficient neuromorphic systems that emulate biological learning. However, conventional synaptic transistors are often incompatible with multiple stimuli or require complex fabrication processes, limiting their integration into scalable hardware. This paper presents an in situ integrated TiO2/SiOx/Al2O3 synaptic phototransistor that supports UV–electrically coupled plasticity and temporal learning dynamics. This sequential in situ process ensures atomically sharp interfaces, and CMOS-compatible scalability, enabling UV–electrically coupled plasticity and temporally adaptive learning dynamics. The device exhibits robust bidirectional conductance tuning under electrical gating and UV exposure and synergistic modulation when stimuli are applied in combination. Temporal pairing of dual stimuli emulates spike-timing-dependent plasticity and enables associative learning, inspired by insect sensory fusion mechanisms involving electrical and optical cues. Furthermore, a system-level evaluation integrating the device's synaptic characteristics into a spiking neural network shows improved robustness to noise under challenging conditions, highlighting the device's potential to emulate biologically inspired selective attention. Our results create a scalable and versatile synaptic platform that bridges material-level integration with system-level functionality, providing a hardware foundation for energy-efficient and noise-resilient intelligent systems.
- Keyword
- associative learning, in situ integration, noise-robust learning, selective attention, UV–electric multimodal synapses
- KSP Keywords
- Adaptive learning, Associative learning, Biologically inspired, CMOS-compatible, Intelligent Systems, Robust Learning, Selective attention, Sensory fusion, Sharp interface, Spike timing dependent plasticity, Spiking Neural Network
This work is distributed under the term of Creative Commons License (CCL)
(CC BY)
(CC BY)
