̽��ֱ�� of Cambridge - Mateja Jamnik

AI: Life in the age of intelligent machines

lw355 — Fri, 22 Feb 2019 14:00:18 +0000

We are said to be standing on the brink of a fourth industrial revolution – one that will see new forms of artificial intelligence (AI) underpinning almost every aspect of our lives. ̽��ֱ��new technologies will help us to tackle some of the greatest challenges that face our world.

In fact AI is already very much part of our daily lives, says Dr Mateja Jamnik, one of the experts who appear in the film. “Clever algorithms are being executed in clever ways all around us... and we are only a decade away from a future where we are able to converse across multiple languages, where doctors will be able to diagnose better, where drivers will be able to drive more safely.”

Ideas around AI “are being dreamt up by thousands of people all over the world – imaginative young people who see a problem and think about how they can solve it using AI… whether it’s recommending a song you’ll like or curing us of cancer,” says Professor Stephen Cave.

Much of the excitement relates to being able to leverage the power of Big Data, says Professor Zoubin Ghahramani. Without AI, how else could we make sense of the vastly complex interconnected systems we now have at our fingertips?

But what do we think about AI and the future it promises? Our perceptions are shaped by our cultural prehistory, stretching right back to Homer, says Dr Sarah Dillon. How we feel about the dawning of a new technology is linked to centuries-old thinking about robotics, automatons and intelligence beyond our own.

And what happens when we come to rely on the tools we are empowering to do these amazing things? Professor Lord Martin Rees reflects on the transition to a future of AI-aided jobs: what will this look like? How will we ensure that the wealth created by AI will benefit wider society and avoid worsening inequality?

Our researchers are asking fundamental questions about the ethics, trust and humanity of AI system design. “It can’t simply be enough for the leading scientists as brilliant as they are to be pushing ahead as quickly as possible,” says Dr Seán Ó hÉigeartaigh. “We also need there to be ongoing conversations and collaborations with the people who are thinking about the ethical impacts of the technology.

“ ̽��ֱ��idea that AI can help us understand ourselves and the universe at a much deeper level is about as far reaching a goal for AI as could be.”

Inset image: read more about our AI research in the ̽��ֱ��'s research magazine; download a pdf; view on Issuu.

In a new film, leading Cambridge ̽��ֱ�� researchers discuss the far-reaching advances offered by artificial intelligence – and consider the consequences of developing systems that think far beyond human abilities.

̽��ֱ��idea that AI can help us understand ourselves and the universe at a much deeper level is about as far reaching a goal for AI as could be

Seán Ó hÉigeartaigh

AI: Humanity's Last Invention?

̽��ֱ��text in this work is licensed under a Creative Commons Attribution 4.0 International License. Images, including our videos, are Copyright © ̽��ֱ�� of Cambridge and licensors/contributors as identified. All rights reserved. We make our image and video content available in a number of ways – as here, on our main website under its Terms and conditions, and on a range of channels including social media that permit your use and sharing of our content under their respective Terms.

Yes

Artificial intelligence is growing up fast: what’s next for thinking machines?

cjb250 — Tue, 06 Feb 2018 09:11:12 +0000

We are well on the way to a world in which many aspects of our daily lives will depend on AI systems.

Within a decade, machines might diagnose patients with the learned expertise of not just one doctor but thousands. They might make judiciary recommendations based on vast datasets of legal decisions and complex regulations. And they will almost certainly know exactly what’s around the corner in autonomous vehicles.

“Machine capabilities are growing,” says Dr Stephen Cave, Executive Director of the Leverhulme Centre for the Future of Intelligence (CFI). “Machines will perform the tasks that we don’t want to: the mundane jobs, the dangerous jobs. And they’ll do the tasks we aren’t capable of – those involving too much data for a human to process, or where the machine is simply faster, better, cheaper.”

Dr Mateja Jamnik, AI expert at the Department of Computer Science and Technology, agrees: “Everything is going in the direction of augmenting human performance – helping humans, cooperating with humans, enabling humans to concentrate on the areas where humans are intrinsically better such as strategy, creativity and empathy.”

Part of the attraction of AI requires that future technologies perform tasks autonomously, without humans needing to monitor activities every step of the way. In other words, machines of the future will need to think for themselves. But, although computers today outperform humans on many tasks, including learning from data and making decisions, they can still trip up on things that are really quite trivial for us.

Take, for instance, working out the formula for the area of a parallelogram. Humans might use a diagram to visualise how cutting off the corners and reassembling it as a rectangle simplifies the problem. Machines, however, may “use calculus or integrate a function. This works, but it’s like using a sledgehammer to crack a nut,” says Jamnik, who was recently appointed Specialist Adviser to the House of Lords Select Committee on AI.

“When I was a child, I was fascinated by the beauty and elegance of mathematical solutions. I wondered how people came up with such intuitive answers. Today, I work with neuroscientists and experimental psychologists to investigate this human ability to reason and think flexibly, and to make computers do the same.”

Jamnik believes that AI systems that can choose so-called heuristic approaches – employing practical, often visual, approaches to problem solving – in a similar way to humans will be an essential component of human-like computers. They will be needed, for instance, so that machines can explain their workings to humans – an important part of the transparency of decision-making that we will require of AI.

With funding from the Engineering and Physical Sciences Research Council and the Leverhulme Trust, she is building systems that have begun to reason like humans through diagrams. Her aim now is to enable them to move flexibly between different “modalities of reasoning”, just as humans have the agility to switch between methods when problem solving.

Being able to model one aspect of human intelligence in computers raises the question of what other aspects would be useful. And in fact how ‘human-like’ would we want AI systems to be? This is what interests Professor José Hernandez-Orallo, from the Universitat Politècnica de València in Spain and Visiting Fellow at the CFI.

“We typically put humans as the ultimate goal of AI because we have an anthropocentric view of intelligence that places humans at the pinnacle of a monolith,” says Hernandez-Orallo. “But human intelligence is just one of many kinds. Certain human skills, such as reasoning, will be important in future systems. But perhaps we want to build systems that ‘fill the gaps that humans cannot reach’, whether it’s AI that thinks in non-human ways or AI that doesn’t think at all.

“I believe that future machines can be more powerful than humans not just because they are faster but because they can have cognitive functionalities that are inherently not human.” This raises a difficulty, says Hernandez-Orallo: “How do we measure the intelligence of the systems that we build? Any definition of intelligence needs to be linked to a way of measuring it, otherwise it’s like trying to define electricity without a way of showing it.”

̽��ֱ��intelligence tests we use today – such as psychometric tests or animal cognition tests – are not suitable for measuring intelligence of a new kind, he explains. Perhaps the most famous test for AI is that devised by 1950s Cambridge computer scientist Alan Turing. To pass the Turing Test, a computer must fool a human into believing it is human. “Turing never meant it as a test of the sort of AI that is becoming possible – apart from anything else, it’s all or nothing and cannot be used to rank AI,” says Hernandez-Orallo.

In his recently published book ̽��ֱ��Measure of all Minds, he argues for the development of “universal tests of intelligence” – those that measure the same skill or capability independently of the subject, whether it’s a robot, a human or an octopus.

His work at the CFI as part of the ‘Kinds of Intelligence’ project, led by Dr Marta Halina, is asking not only what these tests might look like but also how their measurement can be built into the development of AI. Hernandez-Orallo sees a very practical application of such tests: the future job market. “I can imagine a time when universal tests would provide a measure of what’s needed to accomplish a job, whether it’s by a human or a machine.”

Cave is also interested in the impact of AI on future jobs, discussing this in a report on the ethics and governance of AI recently submitted to the House of Lords Select Committee on AI on behalf of researchers at Cambridge, Oxford, Imperial College and the ̽��ֱ�� of California at Berkeley. “AI systems currently remain narrow in their range of abilities by comparison with a human. But the breadth of their capacities is increasing rapidly in ways that will pose new ethical and governance challenges – as well as create new opportunities,” says Cave. “Many of these risks and benefits will be related to the impact these new capacities will have on the economy, and the labour market in particular.”

Hernandez-Orallo adds: “Much has been written about the jobs that will be at risk in the future. This happens every time there is a major shift in the economy. But just as some machines will do tasks that humans currently carry out, other machines will help humans do what they currently cannot – providing enhanced cognitive assistance or replacing lost functions such as memory, hearing or sight.”

Jamnik also sees opportunities in the age of intelligent machines: “As with any revolution, there is change. Yes some jobs will become obsolete. But history tells us that there will be jobs appearing. These will capitalise on inherently human qualities. Others will be jobs that we can’t even conceive of – memory augmentation practitioners, data creators, data bias correctors, and so on. That’s one reason I think this is perhaps the most exciting time in the history of humanity.”

Inset image: read more about our AI research in the ̽��ֱ��'s research magazine; download a pdf; view on Issuu.

Our lives are already enhanced by AI – or at least an AI in its infancy – with technologies using algorithms that help them to learn from our behaviour. As AI grows up and starts to think, not just to learn, we ask how human-like do we want their intelligence to be and what impact will machines have on our jobs?

Perhaps we want to build systems that ‘fill the gaps that humans cannot reach’, whether it’s AI that thinks in non-human ways or AI that doesn’t think at all

José Hernandez-Orallo

̽��ֱ��District

Artificial intelligence

̽��ֱ��text in this work is licensed under a Creative Commons Attribution 4.0 International License. For image use please see separate credits above.

Yes

Preparing for the future: artificial intelligence and us

lw355 — Fri, 02 Feb 2018 09:00:13 +0000

AI systems are now used in everything from the trading of stocks to the setting of house prices; from detecting fraud to translating between languages; from creating our weekly shopping lists to predicting which movies we might enjoy.

This is just the beginning. Soon, AI will be used to advance our understanding of human health through analysis of large datasets, help us discover new drugs and personalise treatments. Self-driving vehicles will transform transportation and allow new paradigms in urban planning. Machines will run our homes more efficiently, make businesses more productive and help predict risks to society.

While some AI systems will outperform human intelligence to augment human decision making, others will carry out repetitive, manual and dangerous tasks to augment human labour. Many of the greatest challenges we face, from understanding and mitigating climate change to quickly identifying and containing disease outbreaks, will be aided by the tools of AI.

What we’ve seen of AI so far is only the leading edge of the revolution to come./system/files/issue_35_research_horizons_new.pdf

Yet the idea of creating machines that think and learn like humans has been around since the 1950s. Why is AI such a hot topic now? And what does Cambridge have to offer?

Three major advances are enabling huge progress in AI research: the availability of masses of data generated by all of us all the time; the power and processing speeds of today’s supercomputers; and the advances that have been made in mathematics and computer science to create sophisticated algorithms that help machines learn.

Unlike in the past when computers were programmed for specific tasks and domains, modern machine learning systems know nothing about the topic in question, they only know about learning: they use huge amounts of data about the world in order to learn from it and to make predictions about future behaviour. They can make sense of complex datasets that are difficult to use and have missing data.

That these advances will provide tremendous benefits is becoming clear. One strand of the UK government’s Industrial Strategy is to put the UK at the forefront of the AI and data revolution. In 2017, a report by PricewaterhouseCoopers described AI as “the biggest commercial opportunity in today’s fast-changing economy”, predicting a 10% increase in the UK’s GDP by 2030 as a result of the applications of AI.

Cambridge ̽��ֱ�� is helping to drive this revolution – and to prepare for it.

Our computer scientists are designing systems that are cybersecure, model human reasoning, interact in affective ways with us, uniquely identify us by our face and give insights into our biological makeup.

Our engineers are building machines that are making decisions under uncertain conditions based on probabilistic estimation of perception and for the best course of action. And they’re building robots that can carry out a series of actions in the physical world – whether it’s for self-driving cars or for picking lettuces.

Our researchers in a multitude of different disciplines are creating innovative applications of AI in areas as diverse as discovering new drugs, overcoming phobias, helping to make police custody decisions and forecasting extreme weather events.

Our philosophers and humanists are asking fundamental questions about the ethics, trust and humanity of AI system design, and the effect that the language of discussion has on the public perception of AI. Together with the work of our engineers and computer scientists, these efforts aim to create AI systems that are trustworthy and transparent in their workings – that do what we want them to do.

All of this is happening in a university research environment and wider ecosystem of start-ups and large companies that facilitates innovative breakthroughs in AI. ̽��ֱ��aim of this truly interdisciplinary approach to research at Cambridge is to invent transformative AI technology that will benefit society at large.

However, transformative advances may carry negative consequences if we do not plan for them carefully on a societal level.

̽��ֱ��fundamental advances that underpin self-driving cars may allow dangerous new weapons on the battlefield. Technologies that automate work may result in livelihoods being eliminated. Algorithms trained on historical data may perpetuate, or even exacerbate, biases and inequalities such as sex- or race-based discrimination. Without careful planning, systems for which large amounts of personal data is essential, such as in healthcare, may undermine privacy.

Engaging with these challenges requires drawing on expertise not just from the sciences, but also from the arts, humanities and social sciences, and requires delving deeply into questions of policy and governance for AI. Cambridge has taken a leading position here too, with the recent establishment of the Leverhulme Centre for the Future of Intelligence and the Centre for the Study of Existential Risk, as well as being one of the founding partners of ̽��ֱ��Alan Turing Institute based in London.

In the longer term, it is not outside the bounds of possibility that we might develop systems able to match or surpass human intelligence in the broader sense. There are some who think that this would change humanity’s place in the world irrevocably, while others look forward to the world a superintelligence might be able to co-create with us.

As the ̽��ֱ�� where the great mathematician Alan Turing was an undergraduate and fellow, it seems entirely fitting that Cambridge’s scholars are exploring questions of such significance to prepare us for the revolution to come. Turing once said: “we can only see a short distance ahead, but we can see plenty there that needs to be done.”

Inset image: read more about our AI research in the ̽��ֱ��'s research magazine; download a pdf; view on Issuu.

Dr Mateja Jamnik (Department of Computer Science and Technology), Dr Seán Ó hÉigeartaigh (Centre for the Study of Existential Risk and the Leverhulme Centre for the Future of Intelligence, CFI), Dr Beth Singler (Faraday Institute for Science and Religion and CFI) and Dr Adrian Weller (Department of Engineering, CFI and ̽��ֱ��Alan Turing Institute).

Today we begin a month-long focus on research related to artificial intelligence. Here, four researchers reflect on the power of a technology to impact nearly every aspect of modern life – and why we need to be ready.

What we’ve seen of AI so far is only the leading edge of the revolution to come.

Mateja Jamnik, Seán Ó hÉigeartaigh, Beth Singler and Adrian Weller

Jonathan Settle / ̽��ֱ�� of Cambridge

̽��ֱ��text in this work is licensed under a Creative Commons Attribution 4.0 International License. For image use please see separate credits above.

Yes

Can machines reason?

bjb42 — Fri, 01 Feb 2008 15:46:42 +0000

Some of the deepest and greatest insights in reasoning have been made using mathematics. It’s not surprising therefore that emulating such powerful reasoning on machines – and particularly the way humans use diagrams to ‘see’ an explanation for mathematical theorems – is one of the aims of artificial intelligence.

Diagrams for reasoning

Drawing pictures and using diagrams to represent a concept is perhaps one of the oldest vehicles of human communication. In mathematics, the use of diagrams to prove theorems has a rich history: as just one example, Pythagoras’ Theorem has yielded several diagrammatic proofs in the 2500 years following his contribution to mathematics, including that of Leonardo Da Vinci’s. These diagrammatic proofs are so clear, elegant and intuitive that with little help even a child can understand them.

̽��ֱ��concept of the ‘mutilated’ checkerboard is another useful demonstration of how intuitive human reasoning can be used to solve problems. If we remove two diagonally opposite corners, can the board still be covered with dominoes (rectangles made out of two squares)? ̽��ֱ��elegant solution is to colour the checkerboard with alternative black and white squares, like the chessboard, and do the same with the dominoes so that a domino is made of one white and one black square. ̽��ֱ��solution then immediately becomes clear: there are more white squares than black squares, and so the mutilated checkerboard cannot be covered with dominoes. This problem is very easy for people to understand, but no system has yet been implemented that can solve it in such an intuitive way.

As these reasoning techniques can be incredibly powerful, wouldn’t it be exciting if a system could learn such diagrammatic operations automatically? So far, few automated systems have attempted to benefit from their power by imitating them. One explanation might be that we don’t yet have a deep understanding of informal techniques and how humans use them in problem solving. To advance the state of the art of automated reasoning systems, some of these informal human reasoning techniques might have to be integrated with the proven successful formal techniques, such as different types of logic.

From intuition to automation

There are two approaches to the difficult problem of automating reasoning. ̽��ֱ��first is cognitive, which aims to devise and experiment with models of human cognition. ̽��ֱ��second is computational – attempting to build computational systems that model part of human reasoning.

Steps along the computational approach are being taken by Dr Mateja Jamnik in the Computer Laboratory with funding for an Advanced Research Fellowship from the Engineering and Physical Sciences Research Council (EPSRC). While at the ̽��ֱ�� of Edinburgh, Dr Jamnik built Diamond, a program that uses diagrammatic reasoning to prove mathematical concepts. However, there are theorems like the mutilated checkerboard that might require a combination of symbolic and diagrammatic reasoning steps to prove them. In Cambridge, Dr Jamnik is now investigating how a system could automatically reason about such ‘heterogeneous’ proofs. This requires combining diagrammatic reasoning in Diamond with symbolic problem solving in an existing state-of-the-art automated theorem prover. ̽��ֱ��way forward is to give heterogeneous reasoning frameworks access to intelligent search facilities in the hope that the system will not only find new and more intuitive solutions to known problems, but perhaps also find new and interesting problems.

Automated diagrammatic reasoning could be the key to making computer reasoning systems more powerful, as well as to providing the necessary tools to study and explore the nature of human reasoning. We might then have a new means to investigate the amazing ability of the human brain to solve problems.

For more information, please contact the author Dr Mateja Jamnik (Mateja.Jamnik@cl.cam.ac.uk) at the Computer Laboratory.

Humans often use diagrams for reasoning, but can computers do the same?

This problem is very easy for people to understand, but no system has yet been implemented that can solve it in such an intuitive way.

Computer Laboratory

Dr Mateja Jamnik

This work is licensed under a Creative Commons Licence. If you use this content on your site please link back to this page.

Yes