Postdoctoral Position (two years) in Statistical Learning with Sparsity
Project description and working tasks
Industrial robots, autonomous cars, stocks trading algorithms, and deep network assisted evaluation of medical images all crucially involve real-time, intelligent and automated decision making from complex and heterogeneous data, at ever growing scale and pace. This presents unprecedented theoretical and algorithmic challenges and opportunities for researchers in intelligently collecting and transforming data into information, predictions and intelligent decisions.
Modern statistical learning is at the forefront of these advances, and the main objective of this postdoctoral position is to develop cutting-edge mathematical and statistical theory and the next generation of computational tools to address the above challenges with an emphasis on learning with sparsity and other emerging data models, and to potentially explore their applications in AI, including medical imaging, automated quality control, and self-driving cars, evaluated on both simulated and real data.
Understanding and exploiting sparsity and other data structures to extract useful information from big datasets with the purpose of making optimal decisions will be the main research thrust for this position. Within this broad framework, the successful candidate is encouraged to develop their own research agenda, in close collaboration with mentors and colleagues. Potential areas of interest include, but are not limited to:
• Learning with multiple structures (such as sparsity and rank constraints)
• Intelligent data sampling and uncertainty quantification in medical imaging
• Time-data trade-offs in learning
• Statistical learning with generative adversarial networks and their geometry,
• Streaming and distributed algorithms for dimensionality reduction (such as sparse principal component analysis)
• Defence against adversarial examples in deep neural nets.