ERC Advanced grant C-SENSE: Exploiting Low Dimensional Models in Sensing, Computation and Signal Processing
Applications are invited for a research associate position to work with Prof Mike Davies, as part of the C-SENSE ERC advanced research programme on Exploiting Low Dimensional Models in Sensing, Computation and Signal Processing. Successful candidates will be based in the Institute for Digital Communications (IDCOM) at the University of Edinburgh, UK. The position should ideally start in January 2020.
Increasingly advanced signal models are being used to inform enhanced computational imaging and sensing solutions. This includes the incorporation of Plug-and-Play (PnP) denoising algorithms as regularisers within optimisation, or the similar exploitation of such methods within large scale approximate Bayesian inference. This also covers the growing trend of the exploitation of deep neural network (DNN) solutions either within such PnP architectures, as stand-alone solutions or as implicit Bayesian priors.
This project will investigate the ability to scale up advanced signal models and approximate Bayesian inference to efficiently solve large scale computational sensing problems but avoiding the restrictive assumptions made by some algorithms, such as fully random forward models. The project will study their ability to deliver meaningful insight into statistical accuracy, even when using complex signal models such as PnP strategies, that can compete with state-of-the-art DNN solutions in terms of computation and performance but without the need for training. Here the notion of computation will be interpreted broadly. Tasks other than full reconstruction with also be considered.
Exemplar applications that could be considered during the project include quantitative MRI and CT imaging, radar, sonar or ultrasound imaging, lidar, low-photon imaging, etc.
Informal enquiries by email can be sent to Prof. Mike Davies (email@example.com).
Salary: £33,797 - £40,322 per annum
Closing date is 28 October 2019 at 5pm GMT