PhD position on Tensor Compression for building Machine Learning Models

The advertised PhD position is at the intersection of research units NUMA and ADVISE at KU Leuven.

The Numerical Analysis and Applied Mathematics research unit NUMA is part of the Computer Science department and works on numerical methods, algorithms and software for simulation and data analysis, with applications in many fields in science and engineering. The research in NUMA focuses, amongst others, on data science, simulation, optimization and high performance computing.

Advanced Integrated Sensing lab (ADVISE) is a research group at the Faculty of Engineering Technology at Campus Geel. ADVISE unites a broad experience in both hardware (design and testing of both PCB and integrated circuit implementations) and software (real-time processing and analysis of large multimodal datasets). This unique combination of expertise enables the development of integrated sensing and communication systems for many different applications including those that need to work in harsh (radiation) conditions or have extremely high reliability requirements.

Project

Context :  Artificial Intelligence has revolutionized the way images are processed. During the last years, spectral imagers (e.g., hyperspectral, CT) have become commercially available, which offer in particular great added for quality control in agrofood. However, spectral imaging devices are burdened by the complexity and high dimensionality of generated data compared to classical imagers. In addition, machine learning models typically require large, labelled datasets which are generally unavailable for spectral images. There is thus a need for efficient learning strategies. The objective of this project is to develop tensor-based compression algorithms for efficiently building machine learning models for high-dimensional, spectral images in agrofood applications.

In this research work, you will first evaluate the performance of tensor decomposition and compression when applied to spectral data and fed into a machine learning pipeline. Subsequently, you will go well beyond the state of the art and develop novel methods that adapt the tensor decomposition to align better with the machine learning goal (e.g., classification, regression, clustering).

This project is in collaboration with the DTAI research unit of the Computer Science department and the MeBioS division of the Biosystems department.

Profile

• You have a Master's degree in Mathematics, Applied Mathematics, Computer Science, Engineering or similar.
• You have experience in programming and a general interest in the development of numerical methods and their application in engineering.
• Excellent proficiency in English is required, both oral and written.

Offer

We offer a full-time position as a doctoralresearcher for a maximum of four years, with yearly evaluation.

Interested?