Hujun Yin is a professor in data science and machine learning at the Department of Electrical and Electronic Engineering at the University of Manchester. He received his PhD degree in neural networks from the University of York, having obtained his MSc degree in signal processing and BEng degree in electronic engineering from Southeast University. He has a wide range of experience in signal/image analysis and data analytics and has had interdisciplinary projects in bioinformatics and neuroinformatic. His recent focus is in image recognition and deep learning with industrial applications.
His research interests range from theories and applications of neural networks, self-organising and deep learning systems in particular; image processing, enhancement and recognition; face recognition; nonstationary signal or time series modelling and prediction; dimensionality reduction and manifold learning. Though his core expertise is in unsupervised and manifold learning, he has developed a variety of methods in a wide range of fields such as gene expressions analysis, protein peptide spectral sequencing, neural signal decoding, signal/image based industrial monitoring, financial time series modelling, robust image feature extraction, as well as hyperspectral image analysis for plant monitoring.
Recently he is particularly interested in solving practical, industrial problems using deep learning frameworks, where unsupervised or data-independent means can be derived for efficient learning. Targeted applications include robust recognition of deformed image objects, imaging inverse problems, and enhanced modelling interpretation for noisy and intermittent signals.