Post-silicon nano-electronic device and its application in brain-inspired chips.

Area of Science:

• Computer Engineering
• Materials Science
• Artificial Intelligence

Background:

• Traditional Von Neumann architecture faces limitations like the 'memory wall' and 'power wall' due to big data demands.
• Brain-inspired computing offers a novel architecture for high energy efficiency and real-time AI performance.
• Memristive devices are key components for creating artificial synapses in neuromorphic computing.

Purpose of the Study:

• To survey post-silicon nano-electronic devices and their application in brain-inspired chips.
• To review current analog and digital brain-inspired chips.
• To analyze the research progress and future prospects of post-silicon devices in neuromorphic computing.

Main Methods:

• Review of neural network development and existing brain-inspired chip designs.
• Analysis of various post-silicon nano-electronic devices.
• Survey of research progress in constructing brain-inspired chips using these devices.

Main Results:

• Identified limitations of traditional computing architectures.
• Detailed the design principles of brain-inspired chips using post-silicon nano-electronic devices.
• Expounded on the research progress of constructing brain-inspired chips using post-silicon nano-electronic devices.

Conclusions:

• Post-silicon nano-electronic devices are crucial for advancing brain-inspired computing.
• Future research should focus on leveraging these devices for next-generation neuromorphic chips.
• Brain-inspired chips offer a promising solution to overcome current computing bottlenecks.