top of page
Research Objectives
Research Objective 1: To develop advanced neuromorphic hardware algorithms adapted to challenges of low power consumption, real-time practical systems, including training mechanisms that go beyond vanilla error back propagation algorithms.
Research Objective 2: To develop novel large scale, low power consumption analogue and neuromorphic computing, Ising machine, reservoir computing, deep recurrent networks, convolutional accelerators and systolic computing hardware. To provide DCs with the corresponding skills in cutting edge technologies.
Research Objective 3: To develop system designs based on the proposed technologies and carry out an extensive experimental study of their performance in practical environments (including field trials). Programmability and reconfigurability of the systems can be enhanced by photonic integrated circuit technologies.
Research Objective 4: To identify new industrial applications with an emphasis on strategic assets (energy efficiency, decentralized computing, speed, communication efficiency). To critically increase economic as well as societal relevance of neuromorphic computing by developing advanced neuromorphic hardware specifically capable of addressing novel yet realistic applications. To ensure smooth transition of scientific innovation to industry through direct involvement of stakeholders in all elements of the chain from academic institutes and research centres to companies.
Research Objective 5: To ensure support of long-term, industry-oriented research that supports society by direct and close collaboration between industry and academia in the field of neuromorphic technology and all-analogue photonic computing, inside and outside of the POSTDIGITAL+ consortium, during and beyond the lifetime of the proposed DN project.
bottom of page