Neuromorphic technologies are also seen as the future of Artificial Intelligence (also a key technology of Topsector ICT). In recent decades, AI has made huge advances, making machines increasingly capable of performing human tasks. Neuromorphic technologies promises to further accelerate this progress, as we can now design electronic systems that work just like the human brain. Chips with a biological, neuromorphic architecture can not only bring exascale computing (a new generation of advanced computing systems) closer, but are also more efficient in terms of space, energy, material use, cost and environmental impact. Neuromorphic technologies are particularly suited to pattern recognition, Machine Learning, creative processes and reasoning. And all in a way that more closely resembles human intelligence (similar to our right brain), allowing information to be processed more efficiently.
Major application areas of neuromorphic technologies
There are a number of promising applications of neuromorphic technologies, including fighting viruses and diseases, brain-computer interfaces, genetic modification, cybernetics, mind augmentation and software neural networks. The application areas of neuromorphic technologies can be broadly divided into three categories:
- Machine Vision
Neuromorphic technologies can significantly improve the performance of visual systems. Machines can understand images and videos, recognise objects, analyse scenes and even detect emotions. This opens the door to innovative applications in autonomous vehicles, surveillance systems, robotics and augmented reality. - Natural Language Processing (NLP)
Neuromorphic technologies allow machines to understand and generate natural language, which has implications for speech recognition, translation services and chatbots, among others. Machines can also better interpret human language and respond appropriately, making communication between humans and machines more fluid. - Neuromorphic engineering
Neuromorphic technologies are also driving the development of new hardware and software architectures. This creates energy-efficient systems that can handle complex tasks. Neuromorphic chips can directly simulate neural networks and are applied for high-speed data analysis, Internet of Things (IoT) and in the medical sector.
Strong starting position for the Netherlands
The Netherlands aims to establish itself as a forerunner in neuromorphic technologies. By cleverly using the principles of the brain and nervous system, it aims to bring about technological advances, significant sustainability benefits and improved competitiveness. The Netherlands' starting position is strong because of its excellent academic reputation in artificial intelligence, neuroscience and computational modelling (simulating and studying complex systems using mathematics, physics and computer science). Many talented researchers and teachers work at Dutch universities and research institutes, doing pioneering work in neuromorphic computing. In addition, the Netherlands has a strong tradition of public-private partnerships in digitalisation and information technology.