Research demonstrates resilient memory endurance and nociceptive capabilities in dual-stacked RRAM, suggesting improvements for artificial intelligence applications.
As data is processed by the von Neumann architecture grows, a bottleneck emerges due to the separation of memory and computation units. Neuromorphic computing, inspired by the human brain, offers a solution by integrating memory and computation in hardware. This study explores ITO/IGZO/ZrO 2 /Ti resistive random access memory (RRAM), based on a simple manufacturing process and high performance, demonstrating stable memory characteristics with an endurance of 10 3 cycles and data retention for 10 4 seconds. The RRAM operates based on oxygen vacancies, enabling linear potentiation, depression, spike‐timing‐dependent plasticity (STDP), and spike‐number‐dependent plasticity (SNDP), essential for adjusting synaptic weights in neuromorphic computing. The device successfully demonstrates the capability to mimic nociceptors by detecting pain through the control of threshold voltage, pulse width, and pulse number, enabling the replication of behaviors such as hyperalgesia and allodynia while processing harmful stimuli in a manner similar to human sensory neurons. These findings highlight the potential of transparent ITO/IGZO/ZrO 2 /ITO RRAM for see through applications in humanoid robots, artificial intelligence (AI) systems, and advanced computing technologies, enabling efficient, brain like processing.
No takes yet. Share an insight, caveat, or question.
Jo et al. (2025) studied this question.
Synapse has enriched 5 closely related papers on similar clinical questions. Consider them for comparative context: