Answers are grounded in this paper and cite their sources.
Demonstrates enhanced image recognition accuracy in neuromorphic visual tasks using a synaptic transistor, suggesting its potential for future applications.
A neuromorphic synaptic transistor based on a Ga2O3/IGZO heterojunction is fabricated using radio frequency magnetron sputtering, demonstrating optical-stimulus-dependent postsynaptic current responses modulated by pulse intensity, duration, and frequency. The device successfully emulates essential biological synaptic behaviors, including short-term plasticity, long-term plasticity, paired-pulse facilitation, and learning–experience behavior. Energy band structure analysis reveals that the energy barrier at the Ga2O3/IGZO interface effectively suppresses the rapid recombination of photogenerated carriers, thereby providing insight into the underlying physical mechanisms governing the device's synaptic behavior. A 3 × 3 synaptic array constructed from the proposed device exhibits distinct learning, memory, and forgetting behaviors. Furthermore, the device is applied to visual preprocessing tasks, achieving significant noise suppression efficiency and enhanced image recognition accuracy under Gaussian noise conditions using a recurrent neural network. These results demonstrate the Ga2O3/IGZO heterojunction synaptic transistor's excellent synaptic plasticity and its promising potential for next-generation neuromorphic vision systems.
No takes yet. Share an insight, caveat, or question.
Ji et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: