Postdoc position under the supervision of Professor Jared Tanner

Machine and Deep Learning have proven to be some of the most effective tools in data science, now reliably surpassing human ability to perform tasks such as classification of images and sophisticated games. The holder of this post will work to advance our understanding of the efficacy and interpretability of these methods, as well as apply these techniques to real world problems such as those of interest to Emirates Airlines. Methods to be explored are likely to include: the scattering transform, models for data which can be rigorously analysed such as deep neural nets with Gaussian weights, dictionary learning, convolutional sparse coding, reversibility, adversarial nets, optimal approximation, function learning, the information bottleneck, and high dimensional geometry. Previous experience in some of the following topics expected to be useful: Machine and deep learning, sparse approximation or signal processing, random matrix theory, and information theory. A successful candidate in this field will be supervised by Professor Jared Tanner within the Numerical Analysis Group.

For more information please visit the following link.