Researchers At UCF Are Developing Brain-Like Devices To Enable Artificial Intelligence (AI) To Work From Anywhere, Without Connecting To The Internet

This article summary is based on the research paper:'MoS2 Synapses with Ultra-low Variability and Their Implementation in Boolean Logic' 

Please don't forget to join our ML Subreddit

Need help in creating ML Research content for your lab/startup? Talk to us [email protected]

AI currently relies on connections to external servers to do the heavy computing, and complex calculations are required to run AI processing or perform unsupervised learning. Current research works are trying to reduce the size and complexity of artificial intelligence circuits. This aims for technologies such as portable, handheld gadgets that can have the circuitry built-in and not require an internet connection. 

This article discusses a new research work conducted by the University of Central Florida to enable artificial intelligence (AI) to function without an internet connection.

This will enable technology ranging from natural languages processing systems such as Siri or Alexa to robotics. Other advanced applications might be used in far-flung corners of the globe or even on other planets. 

The paper “MoS2 Synapses with Ultra-low Variability and Their Implementation in Boolean Logic” mentions the recent discoveries of the novel approach for creating sophisticated electronics. 

Robots now need someone telling them what task they have to perform in an environment. However, the robots will require the capacity of unsupervised learning to adjust to changing settings in the future. 

The new work will enable robots to do numerous tasks such as locating and rescuing someone trapped in a faraway location or continuously monitoring elderly people’s health, and providing evacuation if something goes wrong. 

Because of the tiny footprint and reduced complexity of implementing synapses and neurons, brain-inspired computing enabled by memristors has grown in popularity over the years. High cycle-to-cycle and device-to-device variability have hampered the development of complex neuromorphic circuits utilizing standard materials methods.

Transparent, flexible, ultra-thin memristive synapses for neuromorphic computing have been realized using two-dimensional (2D) materials, albeit with insufficient knowledge of device statistical variance.

The researchers’ intricate neuromorphic — or brain-like —devices are built on microscopic, rectangular chips about an inch broad. Despite the fact that other researchers have attempted to produce similar technologies, the UCF-developed gadgets are more dependable due to their unique design and nanoscale components.

Artificial intelligence will be able to function without having to connect to the internet, thanks to the gadgets that UCF researchers are building.

Instead of growing it on another substrate and then transferring it, UCF researchers developed it directly on the main chip. 

They are manufactured on the same platform, which decreases the chemistry-induced aberrations that occur when the transfer is employed. As a result, they avoided it. The researchers modified the way the current travels through the gadget by employing this alternate approach. This improves device dependability by lowering device variability. 

According to the researchers, the chips might be used in contemporary technologies over the next ten years. Their next steps are to improve the technology, including creating networks with devices that allow new applications such as image recognition.

Paper: https://pubs.acs.org/doi/10.1021/acsnano.1c09904#

Source: https://www.ucf.edu/news/ucf-researchers-create-brain-like-devices-to-enable-ai-in-remote-regions-and-space/