On chip memristive artificial nano synapses and neural networks
These last fifty years have seen Von Neumann computing architectures boom. Nevertheless, even the most powerful digital computers cannot rapidly solve apparently simple problems such as image interpretation. However, because its s...
ver más
¿Tienes un proyecto y buscas un partner? Gracias a nuestro motor inteligente podemos recomendarte los mejores socios y ponerte en contacto con ellos. Te lo explicamos en este video
Proyectos interesantes
SelectX
Integrated Crossbar of Microelectromechanical Selectors and...
125K€
Cerrado
NEURAMORPH
Dynamics of Amorphous Semiconductors Intrinsic Nature and A...
1M€
Cerrado
EUR2022-134045
MEMRISTORES DE PEROVSKITA PARA REDES DE IMPULSOS
90K€
Cerrado
SPIN-ION
Hybrid Spintronic Synapses for Neuromorphic Computing
2M€
Cerrado
RAMP
Real neurons nanoelectronics Architecture with Memristive Pl...
3M€
Cerrado
FREEMIND
Ferroelectric REsistors as Emerging Materials for Innovative...
191K€
Cerrado
Fecha límite de participación
Sin fecha límite de participación.
Descripción del proyecto
These last fifty years have seen Von Neumann computing architectures boom. Nevertheless, even the most powerful digital computers cannot rapidly solve apparently simple problems such as image interpretation. However, because its structure is
massively parallel and analog, the human brain is able to perform such tasks in a fraction of second. Neuromorphic circuits allow to go beyond conventional digital architectures. An on-chip implementation of these circuits requires to be able to fabricate nanometer sized, analog, reconfigurable, fast components. While the spiking neurons can easily be fabricated with classical CMOS technology, the synapse plasticity is challenging to achieve. In 1971 L. Chua has introduced a new circuit element, called memristor , a non-linear resistance which by definition includes a memory effect. Only last year, a team in Hewlett-Packard has for the first time proposed a device for synaptic applications showing memristive properties based on electromigration of oxygen vacancies in Titanium Oxide. The project NanoBrain aims first at developing alternative memristors based on different physical principles (spintronics and ferroelectricity), avoiding in particular the potential over-heating and fragility of the electromigration-based devices. The final goal of the project is to prove the efficiency of these new nano-synapses by integrating them into functional neural networks.