Using biological systems as role models for efficiency
OCT 05, 2012
DOI: 10.1063/PT.5.026412
Nature : In the interest of efficient computation, Susanne Still of the University of Hawaii and colleagues have been studying biological systems that are good at extracting useful information from their environment and doing so with minimal waste of energy. In a paper published in Physical Review Letters, the researchers conclude that the most efficient machines remember what has happened to them and use that memory to predict what the future holds. Parallels can be drawn between the simplest molecular motors and more complicated systems, such as an “intelligent” car. The researchers add that because memory in most systems is finite, the best machines are also highly selective in the information they take in and retain.
© 2012 American Institute of Physics
