Professor Danushka Bollegala - Department of Computer Science, University of Liverpool - 'Gender-preserving Debiasing for Pre-trained Word Embeddings'

October 2019

Word embeddings learnt from massive text collections have demonstrated significant levels of discriminative biases such as gender, racial or ethnic biases, which in turn bias the down-stream NLP applications that use those word embeddings. Taking gender-bias as a working example, we propose a debiasing method that preserves non-discriminative gender-related information, while removing stereotypical discriminative gender biases from pre-trained word embeddings. 


Dr Vanessa Robins -  Department of Applied Mathematics, The Australian National University - 'Insights from analysis of porous and granular materials' 

May 2019

The physical properties of porous and granular materials critically depend on the topological and geometric details of the material micro-structure.  For example, the way water flows through sandstone depends on the connectivity and diameters of its pores, and the balance of forces in a grain silo on the contacts between individual grains.  My work with the x-ray micro-CT group at ANU has produced topologically valid and efficient algorithms for studying and quantifying the intricate structure of complex porous materials.  We have shown that measures such as the Minkowski functionals and persistent homology give clear signatures of crystallisation in bead packs, the degree of consolidation in granular materials, and correlate to physical properties such as fluid permeability and non-wetting phase trapping capacity in sandstones. 


Dr Cinzia Giannetti - Senior lecturer University of Swansea - 'The 4th industrial revolution'

April 2018

Dr Cinzia Giannetti is a senior lecturer in the college of engineering, Swansea University. In this talk, Dr Giannetti gave an overview of enabling technologies underpinning the digital transformation of production processes, also called the 4th industrial revolution (4IR). Among these technologies, she focussed on industrial big data and the challenges that arise when using (big or small) data to optimise manufacturing operations. Through a practical case study, she showed how data can be used to develop predictive models of complex manufacturing systems and discover new insights that can support decision-making capabilities, ultimately leading to increased robustness of manufacturing processes.


Professor Fernanda Irrera - 'Wearing Sensing for the detection of Parkinson's disease'

May 2017

Fernanda Irrera is a professor at the electronics department of the University of Rome “La Sapienza”, will talk about a smart wearable sensing system for the detection of motion symptoms of the Parkinson’s disease. This was an opportunity to hear about wearable sensing systems in action and then decide for yourself whether it is the transformative technology promised.

Martina King - CEO Featurespace - 'Machine learning in action: the Featurespace story'

February 2017

Featurespace is the world’s leading provider of adaptive behavioral analytics technology for fraud and risk management. Martina talked us through how machine learning is changing the way businesses work and how academic research has been translated into successful products.