Microsoft Cloud Computing Research Centre, 3rd Annual Symposium
'Machine Learning: Technology, Law & Policy '
8-9 September 2016
Queens' College, Cambridge
What is Machine Learning (ML) and how does it work in the context of broader workflows that may involve humans, cloud services and the Internet of Things? What is the impact of using ML technologies to process personal data in a European framework and what issues of liability and autonomy arise when decisions are made by machines rather than by humans? These and other issues of vital importance to cloud-supported machine learning were explored at the MCCRC 3rd Annual Symposium in September 2016.
Attendees also had the opportunity to participate in small groups consisting of a balanced number of technical and legal experts and moderated by a representative from Cambridge or QMUL. Group participants elected their own representatives to report the group's findings back to the Plenary Session.
To accompany and facilitate the discussions, the MCCRC research team has produced three academic research papers on the technical, legal and policy considerations for machine learning applications. Please see below for more details.
Symposium Research Papers:
'Responsibility & Machine Learning: Part of a Process'
by Jatinder Singh, Ian Walden, Jon Crowcroft and Jean Bacon
'Machine Learning with Personal Data'
by Dimitra Kamarinou, Christopher Millard and Jatinder Singh
'Responsibility, Autonomy and Accountability: Legal Liability for Machine Learning'
by Chris Reed, Elizabeth J Kennedy and Sara Nogueira Silva
We are also delighted to announce that the Symposium's Keynote presentations are now available on our website. Please find them below:
'Responsibility & ML: Part of a Process' by Jatinder Singh and Ian Walden
Singh and Walden-MCCRC 3rd Annual Symposium_08 Sept 2016.pdf
'Machine Learning with Personal Data' by Christopher Millard and Dimitra Kamarinou
Millard and Kamarinou - MCCRC 3rd Annual Symposium 08 Sept 2016.pdf
'Responsibility, Autonomy and Accountability: Legal Liability for Machine Learning' by Chris Reed
Reed - MCCRC 3rd Annual Symposium 09 Sept 2016.pdf