BIOMACHINELEARNING · Bio-inspired Machine Learning for Chemical Sensing

One of the main obstacles to a wider uptake of general purpose chemical sensing devices, so-called electronic noses (e-noses), are their generally slow response times. Yet, animals use their olfactory senses for essential tasks such as detecting threats and locating food or mating partners. The reason for the enormous success of biological olfactory systems is only partially due to faster sensors. Recent behavioural and physiological work has shown that animals make decisions, and that the response of olfactory brain structures is most informative, long before the receptors and the corresponding receiving neurons in the brain reach a steady state. In our project, we want to exploit a bioinspired approach to enhance e-nose technology and towards enabling their ubiquitous applicability.Our project has three major goals: First, to improve the accuracy and speed of odour detection and identification in electronic nose systems. Second, to provide a bio-inspired solution for odour detection and classification using a spiking neuronal network that outperforms the performance of existing approaches. Third, we aim at implementing our approach on a neuromorphic hardware system, providing a low-power, high-performance computing solution for portable electronic noses. To achieve our goals, we will employ fundamental biological concepts of information processing in neuronal circuits that have been neglected in existing bioinspired systems, but which have been shown to play an important role in biological olfaction.