首页|New Findings in Machine Learning Described from Swiss Federal Institute of Technology (In-sensor Passive Speech Classification With Phononic Metamaterials)
New Findings in Machine Learning Described from Swiss Federal Institute of Technology (In-sensor Passive Speech Classification With Phononic Metamaterials)
扫码查看
点击上方二维码区域，可以放大扫码查看
原文链接
NETL
NSTL
Wiley
Fresh data on Machine Learning are presented in a new report. According to news reporting from Zurich, Switzerland, by NewsRx journalists, research stated, "Mitigating the energy requirements of artificial intelligence requires novel physical substrates for computation. Phononic metamaterials have vanishingly low power dissipation and hence are a prime candidate for green, always-on computers." Funders for this research include European Research Council (ERC), European Research Council (ERC), Horizon Europe Programme. The news correspondents obtained a quote from the research from the Swiss Federal Institute of Technology, "However, their use in machine learning applications has not been explored due to the complexity of their design process. Current phononic metamaterials are restricted to simple geometries (e.g., periodic and tapered) and hence do not possess sufficient expressivity to encode machine learning tasks. A nonperiodic phononic metamaterial, directly from data samples, that can distinguish between pairs of spoken words in the presence of a simple readout nonlinearity is designed and fabricated, hence demonstrating that phononic metamaterials are a viable avenue towards zero-power smart devices. Elastic neural networks composed of phononic metamaterials respond differently to different spoken commands, passively solving a speech classification problem. Their design harnesses the vanishingly low power dissipation of elastic waves, combined with the high expressivity and efficient simulation of metamaterials."
ZurichSwitzerlandEuropeCyborgsEmerging TechnologiesMachine LearningSwiss Federal Institute of Technology
NETL
NSTL
Wiley
