Answers are grounded in this paper and cite their sources.
Development of neuromorphic hardware is the leading route towards reducing giant energy consumption of artificial intelligence infrastructure. Memristor-based chips are among the most promising solutions; however, their fabrication is still limited by laboratory technologies. In this work, we demonstrate an artificial synapse based on Ag/Hf0.5Zr0.5O2 memristor with unique properties. Here, resistive switching is enabled by Ag diffusion into Hf0.5Zr0.5O2 layer during annealing, facilitating the default ON state upon fabrication, thus allowing to bypass electroforming. High reliability together with linear synaptic potentiation/depression and robust STDP demonstration make our memristive synapse a good candidate for the functional element of neuromorphic hardware.
No takes yet. Share an insight, caveat, or question.
Zhidkov et al. (2026) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: