Sommaire:

Novel nanoelectronic devices provide fantastic opportunities for electronic systems, offering new features in an energy efficient way. However, they also possess less desirable features like high variability and low reliability. Interestingly, Biology has to face similar issues, as the brain has to compute exploiting largely imperfect nanodevices. In this talk, I will show how Biology can therefore be an inspiration for novel circuits that leverage the physics of nanoelectronics, with examples based on three physical systems:
- emerging memory devices ("memristors") used as synapses in learning systems
- stochastic nanodevices from spin electronics used for low power computation
- stochastic resonance-based nanocircuits that work by exploiting thermal noise.

I will show that these ideas could allow systems capable of complex cognitive tasks in a power efficient way. They could thus ideally complement traditional microelctronics, for the tasks where it is the least efficient.

---------

Pour plus d'informations, merci de contacter Metz R.