Dr. Jim Herd
Jim Herd is a lecturer in Electronic Engineering at Heriot Watt University. After completing his degree and PhD he spent many happy years designing instrumentation for astronomers. In the 1980s he became Technical Director for Robot Technology Ltd, but subsequently returned to academia to become involved with research into multi-robot workcells with IBM. In the early 2000 he started to devote more and more of his time to developing a robotics programme for primary school children. This culminated in the Robokid program which ran for 9 years until 2016.
Jim will talk about the role of robotics in enthusing children and students, of all ages, in the richness of science and technology. He will illustrate his talk with examples from his own work.
Amanda is a second-year PhD student at Heriot-Watt University. Her research focuses on developing and evaluating conversational systems like Siri and Google Now. Over the past year, she has been working as part of a team on the Amazon Alexa Challenge, a university competition to develop a system able to converse about many topics such as a movies, music and entertainment. Their system, Alana, is one of the three systems elected to take part in the finals of the challenge.
Prof. Mike Chantler
Mike Chantler has a first in Electrical & Electronic Engineering from Glasgow and PhD in image processing from Heriot-Watt. For a while after graduation he was a software engineer on mainly military related projects and did consultancy in Europe before taking up a post initially in Heriot-Watt’s Electrical Engineering Department.
He has successfully supervised over twenty Phds, published over 100 peer-reviewed papers, and attracted multi-million pound EU and national funding. He has been a member of a considerable number of EPSRC, ESRC, Data Lab and InnovateUK panels and has been involved in strategy development for ESRC, EPSRC and InnovateUK (for instance he has been a member of thje EPSRC ICT SAT and DE’s PAB).
He is interested in using machine learning to extract meaning and value from data and presenting this to users in a way that promotes understanding and trust. This not only means being transparent regarding the provenance of the data, but also making the reasoning accessible and understandable to users who may be domain experts but have little or no knowledge of machine learning algorithms. Without trust there is little point to performing the inferencing as users and stakeholders will not act on the results no matter how interesting they might first appear.