Microsoft Research-Cambridge University Machine Learning Initiative
Microsoft (opens in new tab)‘s partnership with the University of Cambridge (opens in new tab) aims to boost the number of AI researchers in the UK and help them change the world for the better. The Microsoft Research-Cambridge University Machine Learning Initiative is supporting PhD students, postdoctoral researchers and interns at the world-leading university, working closely with Microsoft Research (opens in new tab). This effort is supported by the Engineering and Physical Science Research Council Prosperity Partnership entitled “Machine Learning for Tomorrow: Efficient, Flexible, Robust and Automated (opens in new tab)“
Our aim is to realise artificial intelligence’s potential in enhancing the human experience and to nurture the next generation of researchers and talent in the field.
“Cambridge has a culture of ideas going back and forth between industry and academia, and this agreement with Microsoft is a prime example. By working together with industry on issues such as how best to use AI and machine learning, we can not only help solve complex issues for industry, but continue to support world-leading research and train the next generation of leaders in the field.”
Professor Andy Neely, Pro-Vice-Chancellor for Enterprise and Business Relations at Cambridge
Microsoft is working hard on the biggest challenges in the field of AI so we can develop tools that will change lives and organizations across the world. Skype Translator (opens in new tab) has helped people communicate in many languages and countries. After 10 years of research we’ve released Infer.NET (opens in new tab), a cross-platform framework for model-based machine learning that is used in the TrueSkill ranking system (opens in new tab) for Xbox Live (opens in new tab), as well as asthma research and gene analysis, among many other areas.
This initiative aims to address some of the fundamental limitations of current AI technology. For example, existing AI approaches are extremely data-hungry often requiring millions of carefully labelled examples for training. These methods can be fragile, with changes to the input that should have no effect leading to radically different predictions, which can be confidently wrong. Truly data-efficient, robust AI approaches are required in many key areas, including healthcare, personalisation, and resource allocation. The goal of this partnership is to develop the AI tools of tomorrow. R
“Through this initiative, we are bolstering the University of Cambridge’s AI research capacity and capability, thereby increasing the flow of people and ideas between the Microsoft Cambridge lab and the university. Staff at the University of Cambridge are already lecturing in the Microsoft AI Residency Program and contributing to major industry-led projects. Microsoft researchers teach at the university and supervise projects at all levels, and this initiative will further increase that exchange of knowledge.”
Professor Christopher M. Bishop, Laboratory Director, Microsoft Research Cambridge
The parts of society that will benefit most from machine learning will need sophisticated solutions that reflect the complexities of the world in which we live. This partnership is an example of how we are fostering close collaborations with academia to help increase the capacity and capability of machine learning research in universities around the world. This EPRSC Prosperity Partnership programme is aligned with the UK AI Sector Deal.