Descripción del proyecto
Las redes neuronales podrían funcionar pronto en «hardware» cuántico
Las redes neuronales artificiales, que simulan el modo en que el cerebro humano analiza y procesa la información, se utilizan para modelizar patrones complejos y problemas de predicción. Este método suele implicar el desarrollo de «software» en lugar de la creación de «hardware» que imite a las neuronas. El proyecto Quromorphic, financiado con fondos europeos, prevé implantar la computación neuromórfica a nivel de «hardware». Su objetivo es construir el primer ordenador de redes neuronales dedicado que funciona según los principios de la mecánica cuántica. Se construirá en un «hardware» hecho de circuitos eléctricos superconductores. El «hardware» cuántico neuromórfico podría superar a las arquitecturas clásicas de von Neumann, ya que puede entrenarse con múltiples lotes de datos reales en paralelo.
Objetivo
The Quromorphic project will introduce human brain inspired hardware with quantum functionalities: It will build superconducting quantum neural networks to develop dedicated, neuromorphic quantum machine learning hardware, which can, in its next generation, outperform classical von Neumann architectures. This breakthrough will combine two cutting edge developments in information processing, machine learning and quantum computing, into a radically new technology. In contrast to established machine learning approaches that emulate neural function in software on conventional von Neumann hardware, neuromorphic quantum hardware can offer a significant advantage as it can b e trained on multiple batches of real world data in parallel. This feature is expected to lead to a quantum advantage. Moreover, our approach of implementing neuromorphic quantum hardware is very promising since there exist indications that a quantum advantage in machine learning can already be achieved with moderate fault tolerance. In a longer term perspective neuromorphic hardware architectures will become extremely important in both, classical and quantum computing, particularly for distributed and embedded computing tasks, where the vast scaling of existing architectures does not provide a long-term solution. Quromorphic aims to provide proof of concept demonstrations of this new technology and a roadmap for the path towards its exploitation. To achieve this breakthrough, we will implement two classes of quantum neural networks that have immediate applications in quantum machine learning, feed forward networks and non-equilibrium quantum annealers. This effort will be completed by the development of strategies for scaling the devices to the threshold where they will surpass the capabilities of existing machine learning technology and achieve quantum advantage. In preparation for future exploitation of this new technology, we will run simulations to explore its application to real world problems.
Ámbito científico
- natural sciencescomputer and information sciencessoftware
- engineering and technologyelectrical engineering, electronic engineering, information engineeringelectronic engineeringcomputer hardwarequantum computers
- natural sciencescomputer and information sciencesartificial intelligencemachine learning
- natural sciencescomputer and information sciencesdata sciencedata processing
- natural sciencescomputer and information sciencesartificial intelligencecomputational intelligence
Palabras clave
Programa(s)
Convocatoria de propuestas
Consulte otros proyectos de esta convocatoriaConvocatoria de subcontratación
H2020-FETOPEN-2018-2019-2020-01
Régimen de financiación
RIA - Research and Innovation actionCoordinador
91054 Erlangen
Alemania