Offer DescriptionRobust spiking neural network to monitor marine biodiversity (large cetaceans)Preserving biodiversity is fundamental to the survival of species. In the context of underwater fauna, monitoring the activity of large cetaceans (sperm whales, bottlenose dolphins, etc.) is essential for multiple reasons: (i) firstly, species preservation, (ii) secondly, as large cetaceans are at the "end" of the food chain, their depletion results from a depletion of fishery resources, and consequently a direct economic impact on the fishing industry. Accessing and processing natural data (sounds emitted) from large cetaceans is also relevant to better understanding their individual and/or group behaviors (cognitive capacities, learning, communication between individuals) as well as locating them in deep waters.Large cetaceans - such as sperm whales - emit clicks (acoustic signatures) which they use to communicate with each other when hunting giant squid in cohorts (they also use clicks to locate their prey, thanks to a "sonar" technique). These clicks can be picked up by hydrophones and then analyzed (for classification, counting individuals, etc.). To this end, and in the interests of energy efficiency, the aim of this post-doc is to simulate Spiking Neural Networks (SNNs). The input layer of these SNNs will receive natural data in the form of electrical pulses (spikes), which encode the information conveyed by the acoustic signal in frequency and time. Initially, the SNNs will be simulated using open-source simulators; the tools developed will be evaluated in terms of their classification robustness. Secondly, the implementation of these SNNs on low-power electronic chips will be considered, with particular attention to figures of merit such as power consumption, execution speed and footprint.This 18-month Post-Doc project within the CRIStAL laboratory is part of a wider ANR project "ULP SMART 3D COCHLEA" (partners are CRISTAL and IEMN - Univ. Lille - and INPS - Univ. Toulon -) which started on January 1, 2022, and will run for 4 years. The post-doc will be based at IRCICA.Please send a CV and a motivation letter to to apply.RequirementsAdditional InformationWebsite for additional job detailsWork Location(s)Number of offers available 1 Company/Institute CRIStAL (Research Center in Computer Science, Signal and Automatic Control of Lille) Country France City Lille GeofieldWhere to apply E-mailPierre.Boulet@univ-lille.frContact CityVilleneuve d'Ascq WebsiteSTATUS: EXPIRED