Computational Creativity: Deep Learning for Computer Generation and Manipulation of Photographs, Videos, 3D

Description

The goal of this research is to develop novel methodologies enabling generative modelling of audiovisual content. Such generative models actually enable a computer to create new content by sampling the model. Recent publications in the machine learning literature have shown impressive results in generating sounds, photographs, or even computer art. The focus here will be on making such approaches more useful to artists and creative people by developing novel approaches enabling to parametrically control the generated or manipulated end-result. Different novel machine learning approaches will be investigated: neural Turing machines, HMMs, adversarial networks, reinforcement learning, wide and deep models, multidimensional embeddings, sequence to sequence mapping, recurrent neural network. The research will allow the candidate to develop a unique and transferable expertise in advanced machine learning, AI and Big Data. The researcher will be integrated within a significant group of deep learning researchers involved in regional and international collaboration with world-leading labs in the area.