Jakarta, INTI - Researchers from Loughborough University in the United Kingdom have successfully created a new semiconductor chip that is inspired by the way the human brain works. This innovation has the potential to make artificial intelligence (AI) systems much more energy efficient, up to 2,000 times more efficient at certain tasks.
Unlike conventional AI systems that rely on software, this chip is capable of processing time-varying data directly through its hardware. This approach enables faster processing while significantly reducing energy consumption.
The research was led by Pavel Borisov with support from the Engineering and Physical Sciences Research Council. According to Borisov, this research opens up new ways to design more efficient AI systems.
The Chip is Developed Mimicing Human Brain
The chip utilizes a special component called a niobium oxide memristor, which has a porous and random microstructure. This design mimics the complex neural networks of the human brain, enabling it to perform tasks, such as simple logic operations, image recognition, and time-based data prediction.
In tests using the mathematical model of the Lorenz-63 system, known for describing the “butterfly effect” phenomenon, the chip was able to predict short-term patterns and reconstruct missing data with a high degree of accuracy.
Furthermore, the chip successfully recognized numbers from pixelated images and performed basic logic operations, demonstrating that a single device can handle multiple AI functions simultaneously.
The Research Resulted More Energy Efficiency
The main advantage of this technology lies in its energy efficiency. By mimicking the way the brain builds complex and random neural connections, the researchers have developed a system that is not only intelligent but also extremely energy-efficient.
In the future, this technology has the potential to become the foundation for the development of neuromorphic electronics, devices designed to mimic the brain's workings and capable of efficiently processing time-based signals for a variety of modern AI applications.
Conclusion
Researchers from Loughborough University have developed a semiconductor chip that mimics the workings of the human brain, allowing it to process data directly in hardware, making AI systems significantly faster and up to 2,000 times more energy efficient. Led by Pavel Borisov, the research performed tasks such as data prediction, image recognition, and logic operations, including in tests using the Lorenz-63 system model.
Read more: AI Fuels Strong Growth for Nvidia and iPhone Related Chip Vendors