The university will create human AI research teams to strengthen research

0
© iStock/Ryzhi

New research funding will support the University of Manchester in its drive to create human AI research teams to meet future health needs.

The University of Manchester is teaming up with artificial intelligence (AI) researchers to try to overcome the limitations of current AI systems.

Professor Sami Kaski, from the University of Manchester, has been named among the first Turing Artificial Intelligence (AI) Research Fellows. With his fellowship, Professor Kaski aims to overcome the challenge attached to AI systems that require detailed specification of the goal before they can help.

Machine learning, a form of AI, automatically learns solutions to problems from data. This includes healthcare, where AI can detect patterns associated with diseases and health conditions by studying medical records and other data.

For healthcare applications, the AI ​​business will build on the multimillion-pound Christabel Pankhurst Institute for Health Technology, which aims to capitalize on the University’s strengths in digital health, AI and advanced materials and to develop innovative products and services for the health sector. . This, in turn, will stimulate business growth and employment, as well as the long-term health benefits of the urban region.

As part of this new AI-driven approach, the University of Manchester has also received a share of UKRI’s £4.4m research funding, in addition to partner and community contributions. university worth over £10 million.

AI limits

Artificial intelligence is limited by the fact that human intervention is required to set appropriate goals and rewards to tell AI systems what outcomes are desired. This is difficult when the goal is only partially known, as is the case at the beginning of scientific research.

In drug design, for example, today’s most advanced tools are able to generate candidate molecules if we can specify a precise objective function for them. However, this is difficult for humans to do – and if our specification is not perfect, the system will produce undesirable results. The researchers say this is where the new approaches will come in handy.

The potential of AI in research and complex decision-making is still relatively untapped. Now Professor Kaski aims to develop new ways for machine learning systems to help humans in the process of designing experiments and interpreting the meaning of the results, before deciding what to measure next, and finally find reliable solutions to problems. In personalized lung cancer medicine, for example, to maximize the effectiveness of radiation therapy for a new patient while minimizing side effects, physicians must combine their expertise with what can be learned from measurements of previous patients.

Professor Kaski said: “This is where AI can help, but we need new types of AI assistants that can learn to work well with humans and complement their skills. This requires new fundamental research on AI, and I’m glad Manchester has recognized this opportunity and is significantly stepping up its AI research.

This new approach will be applied to three challenges: diagnosis and therapeutic decision in personalized medicine; guiding scientific experiments in synthetic biology and drug design; and the design and use of digital twins to design physical systems and processes.

Digital twins, a virtual representation of complex objects or systems, can be constructed for patients for personalized medicine, but also for physical systems, such as complex buildings, a farm, and even a city. With the twins, it is for example possible to plan road changes and anticipate the effects on traffic and air quality.

An AI Center of Excellence will be established at the University of Manchester, in collaboration with the Turing Institute and a number of industry and healthcare partners, and with strong links to networks of top researchers national and international in AI.




Subscribe to our newsletter

Recommended related articles





Share.

Comments are closed.