Artificial Intelligence

Dr Franziska Kohlt (Oxford University) presented a History of Ideas lecture,

‘Fairylands of Science: The Alice in Wonderland World of Artificial Intelligence’.

 This was an introduction to AI by thinking about it in a slightly different way. Dr Kohlt approached AI as a subject of scientific research by exploring ‘the scientific imagination…considering the Arts and Humanities – the greatest scientific minds of all time have always done it that way.’

 

Dr Franziska Kohlt is a researcher in literature, the history and philosophy of science and communication and media science. She completed her doctorate at the University of Oxford. Her thesis investigated the mutually influential relationship between science and literature in the emergence of the psychological sciences in the 19th century.

Dr Kohlt regularly appears at international conferences, radio and television as an expert on Lewis Carroll and Alice in Wonderland. She has most recently been working on a panel on “AI and Comedy” alongside Jimmy Carr and the makers of “Have I Got News For You”.

Emma Lacina-Moser 6M4

Dr Kohlt’s lecture was an engaging, relevant and holistic approach to the topic of Artificial Intelligence. As a humanities student, my interest in this subject was shrouded with doubts that I would be able to keep up with the technicalities typically associated with such a topic. My concerns were wholly unnecessary since Dr Kohlt’s talk perfectly synched the psychological, sociological with the logical. The inspired focus on Lewis Carroll’s Alice in Wonderland was a satisfying link from which one could delve deeper into the work’s implications in the sphere of AI. I felt comfortable and well-led through the entirety of this far-stretching theme that was deeply encrypted in contemporary science/psyche. We are all very grateful for Dr Askey for inviting Dr Kohlt who gave us plenty of food for thought.

Dr Kohlt also ran an introduction to AI workshop for yr7s. Here are some views:

Thomas Marsh 7H

It was a brilliant opportunity and a unique experience to be able to discuss and learn about AI and take part in several group activities. However much we already knew about AI, we all walked away with our brains a lot fuller than before. We learnt about a range of aspects of AI, from its history to its future. Everyone felt free to share their opinions and thoughts. I just hope they enjoyed the experience as much as I did. I loved every second of it and I feel lucky to have had the opportunity to share my thoughts and learn many things that I didn’t know before. Thank you.

Mateo Jalba 7B

This experience was so intriguing! We learnt that one needs engineers, designers and even psychologists and synthetic biologists to successfully build an autonomous car- it was incredible! Two hours was not enough to discover the whole new world of Artificial Intelligence!

Michael Worsley 7Y

At the beginning of the workshop, we learnt what AI is. We then looked at some current AI robots and mythological automatons, like PARO and TALOS and did some role play on what society’s views might be on them. Finally, we were asked a philosophical question how we would design an autonomous car, which employs AI, to make appropriately moral decisions. Finally, we did some marketing of our autonomous cars – how we would advertise its moral decision-making and safety features. Thank  you Dr Kohlt!

 

The Langton at the Vanguard

Teaching Artificial Intelligence

Dr E Askey

The headlines read ‘Mona Lisa brought to life’. This is achieved by Deepfake technology from Samsung’s AI research laboratory in Moscow. Samsung’s AI system mapped facial features and movements from a public database of celebrities gathered from YouTube to anticipate what Mona Lisa looked like in real life. Whilst this is highly impressive and entertaining, it has a disturbing potential for misuse. This technology can be used to literally put words into people’s mouths. A fake video of President Barack Obama was created by researchers at Tel Aviv University in 2017. If not regulated, there is the potential danger of never knowing what is, and what is not, fake news.

We might say – well what has this got to do with my everyday life? Well, if you have a supermarket loyalty card, it’s got everything to do with you. The supermarket in question is watching your every purchase and is using AI to make assumptions about you and your buying preferences. Its algorithms are learning how to entice you to buy more.

Langton students will have jobs in the future that are yet to be invented. There is the likelihood that they will need to be well versed in the use and understanding of AI. So we are preparing them for their future.

A group of 12 yr12 students are working with me to learn about the present state of AI research and use. We are aiming to use this information to construct a Key Stage 3 course in Introduction to AI to be rolled out next academic year.

The following describes our present focus of research.

Medical Applications of Machine Learning by Edward Levings (6B4) & Josh Greening (6B4)

AI is becoming increasing used in identification of early stages of diseases and epidemics because of its ability to offer, reliable and objective analysis of patterns in patient data. With increasing use of deep learning, such data analysis has been successfully applied in Ophthalmology, a branch of medicine concerned with the diagnosis, treatment and prevention of diseases of the eye. Machine learning applications in medicine is likely to increase in other areas of medicine where human surveys of patient data ca be replicated by computer analysis. But the ethics of AI indicate that expert human collaboration with AI will always be essential to ensure anomalies can still be identified with confidence.

Manufacturing and AI by Alfie Drew (6H6) & Safin Gurung (6M1)

Machine learning is deeply integrated application of artificial intelligence that is already present in day to day manufacturing. The concept behind machine learning in manufacturing is to lower costs whilst improving production quality. For machine learning to be fully capable of manufacturing skillsets, algorithms will have to be fed large amounts of data to gain intelligence on anomalies and quality control standards. AI is capable of understanding these concepts and integrating them into the manufacturing process.

Cyber Security with AI by Alex Dickers (6S6)

I am currently researching how cyber security needs to adapt in order to keep machine learning and AI systems and databases secure. I am also researching how AI and machine learning can be used in cyber security to improve the security of personal data as well as mitigating malicious and non-malicious cyber-attacks. This will give The Langton an overall view as to what the future of cyber security might look like.

Applications of Machine Learning in Physics Research by Dan O’Reilly (6S4) and Lily Stollery (6H3)

We are researching the use of AI in Physics Research. A recent ground breaking moment was at CERN in July 2012 where they announced the discovery of the Higgs Boson, an elementary particle which is incredibly important to the Standard Model of particle physics. Machine learning played a role in helping to detect the pattern of decay which was required to find the Higgs Boson.

Another sector of physics in which machine learning has a wide variety of applications is astronomical research. An example of this is the Keppler mission which seeks to discover Earth-sized planets orbiting other stars which collects data from the Orion Spur and machine learning is used to find the interesting and outstanding parts of data.

AI in Education Provision by Owen Court (6H2) and Palin Stanley-Webb (6M3)

 Is this technological improvement thanks to Machine Learning!”

As you are probably aware, students learn in vastly different ways. Machine Learning Technology has not neglected this part of society. In our research we are looking at the personalisation of education: how to keep struggling students up to speed, while simultaneously giving top-of-the-class students a new challenge every day. Additionally, we are looking at making teachers’ jobs easier with unbiased, automated marking systems, and even an AI that writes the curriculum for us using predicted jobs in coming years. Writing teacher evaluations, and using these evaluations to match teachers to certain students based on the way they learn, and revolutionary “closed experimentation” on parts of the education system itself. We are even looking at whether education will be needed at all in the future!

All in all, education is a substantial topic regarding Machine Learning in the future, and will be a very crucial area on telling whether machine learning will be a boon or hindrance to our society in the future.

AI and Crime Prevention by Jagger Garcia (6M3) and Josh Matheson (6B2)

We are analysing varying aspects of the applications of Machine Learning in modern society whilst bearing in mind the implications and importance of how data manipulation and AI could impact our society in the future. Our initial research is focussed around law enforcement and the legal sector. At present, AI is regularly used in geographical prediction of crime occurrence and facial recognition technology with a view to developing crime prevention strategies and reconstructing facial images in forensic analysis.

Future AI research areas in the coming months:

  • Politics
  • Ethics/Legalities
  • Climate change/environmental studies
  • Research of Antiquities
  • Development of Genetically modified crops
  • Warfare
  • Transport
  • Financial Sector

Further Reading:

Tom Chivers, The AI Does Not Hate You Superintelligence, Rationality and the Race to Save the World. (Weidenfeld & Nicolson, 2019).

Marcus Du Sautoy, The Creativity Code: How AI is learning to write, paint and think. (Harper Collins, 2019).

Hannah Fry, Hello World: Being Human in the Age of Algorithms. (W. W. Norton Company, 2019).

Stuart Russell, Human Compatible: AI and the Problem of Control. (Viking, 2019).