With the growing pace of development, researchers have been experimenting with various artificial intelligence technologies through devices and products. However, the implementation of artificial intelligence in electronic devices requires custom hardware, which in turn leads to high power consumption. It has been a challenge for researchers to solve problems related to the high power consumption that limits the integration of AI and electronic devices. In a recent breakthrough, scientists at the Korea Advanced Institute of Science and Technology (KAIST) have developed a new artificial intelligence system based on human brain activity.
In this development, the researchers were inspired by the brain’s ability to perform neuromodulation, also known as the “stashing system”. They propose an AI network that constantly changes depending on the situation. The study was published in the journal Advanced Functional Materials and supported by KAIST, the National Research Foundation of Korea, the National NanoFab Center and SK Hynix.
The brain has the ability to transform its neural topology as needed, allowing it to store or recall memories as needed. The researchers were inspired by this brain property, which they implemented by using neural coordination circuit configurations in the AI learning method.
The team, led by Professor Kyung Min Kim from the Department of Materials Science and Engineering at KAIST, developed the hardware technology that can efficiently sustain AI mathematical operations by mimicking the brain through continuous changes in the neural network topology.
“In this study, we implemented the human brain’s learning method with just a simple circuit composition, and as a result, we were able to reduce the power consumption by almost 40 percent,” Professor Kim said in an official statement. The new hardware was able to reduce power consumption by 37 percent by using the stashing system and suffered no loss of accuracy.
The stashing system used by the researchers in the hardware consists of a self-rectifying synaptic array and an algorithm. The system is also compatible with existing electronic devices and semiconductor hardware currently on the market.
https://gadgets360.com/science/news/researchers-propose-new-energy-efficient-ai-hardware-technology-inspired-brain-neuromodulation-korea-advanced-institute-of-science-and-technology-2993224 Researchers propose energy-efficient AI hardware technology inspired by the brain’s neuromodulation capacity