Eco-friendly computing

Kyoto, Japan — The development of neural networks or AI tools for data analysis is increasing exponentially. However, networks that exist in natural ecosystems, such as webs of interspecies relationships, have information processing potential that remains largely untapped.

Now, a study conducted at Kyoto University has shown ecosystem computing power, providing a new direction for the fast-growing AI technology. Simulation has ensured that ecological networks, such as prey-predator interactions, can efficiently process information and use it as a computing resource.

“We have named this approach ecological reservoir computingsaid Kyoto University lead author Masayuki Ushio.

The researchers developed two types of computational ecological reservoirs as a proof of concept that ecological networks have computational power.

One type is the so-called computer-based approach in silicon ecological reservoir computing, which models the dynamics of a hypothetical ecosystem and simulates the system response. The second is the so-called empirical system real-time ecological reservoir computingwhich uses real-time population dynamics of unicellular organisms Tetrahymena thermophila.

In the second approach, to ascertain the computational power of natural ecological systems, Ushio’s team created an experimental design using Tetrahymena thermophila. After entering the value as the culture medium temperature — or input data — the team obtained the cell number as the system output. This study confirms that possibility Tetrahymena populations can make predictions of ecological time series in the near future.

“Our results also suggest that there may be a link between high biodiversity and high computing power, shedding light on new values ​​of previously unknown biodiversity,” added Ushio, currently a principal investigator at the Hong Kong University of Science and Technology. .

“Direct links between community diversity and computational ability can enhance biodiversity intelligence.”

Ecological communities process large amounts of information in real time in natural ecosystems, where the potential for ecological interactions to serve as new computational methods is very high.

“Our new computing method may lead to the invention of a new type of computer. In addition, in developing ways to measure the information processing capacity of a natural ecosystem, we can find clues about how ecosystem dynamics are maintained,” concluded Ushio.

###

The paper “Computational capabilities of ecological dynamics” appeared on April 19, 2023 in Royal Society Open Sciencewith doi: 10.1098/rsos.221614

About Kyoto University

Kyoto University is one of Japan and Asia’s premier research institutions, founded in 1897 and responsible for producing many Nobel laureates and winners of other prestigious international prizes. A broad curriculum across the arts and sciences at undergraduate and graduate levels is complemented by numerous research centers, facilities and offices throughout Japan and the world. For more information, please see: http://www.kyoto-u.ac.jp/en