Computers capable of learning and processing information more efficiently in terms of energy consumption, inspired by the way the human brain works. This is the prospect emerging from a study published in Nature Electronics involving the University of Pisa, Shanghai University, the University of Messina and the École Polytechnique Fédérale de Lausanne (EPFL), coordinated by Professor Andras Kis.
“The human brain can process enormous amounts of information while consuming only a fraction of the energy required by artificial intelligence systems,” explains Giuseppe Iannaccone, Professor at the Department of Information Engineering of the University of Pisa and co-author of the study. “One reason is that, in the brain, memory and processing functions are tightly integrated within the same neural networks. In computers, by contrast, data must be continuously transferred between memory and the processor, resulting in significant energy consumption.”
To move closer to the biological model, the researchers developed new electronic memories based on ultra-thin two-dimensional materials that make it possible to store information and perform computing operations in the same location on the chip. This result was achieved through a new architecture combining two materials: molybdenum disulphide and niobium disulphide. This combination reduces one of the main limitations of two-dimensional devices, namely contact resistance, significantly improving performance.
“A further advantage is that this new technology is compatible with the silicon chips currently produced by the electronics industry,” adds Giuseppe Iannaccone. “The current challenge is to make artificial intelligence increasingly powerful while also making it more energy-efficient and suitable for use in small-scale sensors and autonomous devices.”
The research forms part of the work initiated by the European project QUEFORMAL, coordinated by Iannaccone and dedicated to the development of innovative electronic devices based on two-dimensional materials.
