Published Date: 15/06/2024
The rapid growth of artificial intelligence has led to a significant increase in energy consumption. In fact, projections suggest that AI will account for half a percent of global energy consumption by 2027, equivalent to the energy used by the entire country of the Netherlands. To address this issue, researchers have been working on developing new AI chips that can improve energy efficiency.
One such researcher is Sieun Chae, an assistant professor of electrical engineering and computer science at Oregon State University. Chae has been researching chips based on a novel material platform that allows for both computation and data storage, mimicking the way biological neural networks handle information storage and processing.
Chae's research, published in Nature Electronics, focuses on chips featuring components called memristors, short for memory resistors. Memristors are similar to biological neural networks in that neither has an external memory source, reducing energy loss. By optimizing the composition of memristors, Chae's team has been able to develop chips that can perform tasks with far less energy than a computer's central processing unit.
The potential benefits of this technology are vast. Not only could it reduce the energy footprint of AI, but it could also enable artificial neural networks to process time-dependent data, such as audio and video. This could lead to significant advancements in AI applications across various industries.
The study was funded by the National Science Foundation and involved researchers from the University of Michigan, University of Oklahoma, Cornell University, and Pennsylvania State University.
Oregon State University College of Engineering is a leading institution in engineering research and education. The college is committed to advancing the field of artificial intelligence and reducing its energy footprint.
Oregon State University College of Engineering is a premier institution for engineering research and education. With a strong focus on innovation and sustainability, the college is dedicated to advancing the field of artificial intelligence and reducing its environmental impact.
Q: What is the current energy consumption of artificial intelligence?
A: Projections show that artificial intelligence will account for half a percent of global energy consumption by 2027, equivalent to the energy used by the entire country of the Netherlands.
Q: What are memristors?
A: Memristors are components that can store data and perform computations, similar to biological neural networks. They can significantly reduce energy consumption in AI applications.
Q: How do memristors reduce energy consumption?
A: Memristors reduce energy consumption by eliminating the need for external memory sources, thus minimizing energy loss due to data transfer.
Q: What are the potential applications of this technology?
A: The potential applications of this technology include reducing the energy footprint of AI, enabling artificial neural networks to process time-dependent data, and advancing AI applications across various industries.
Q: Who funded the study?
A: The study was funded by the National Science Foundation.