Preamble: In October 2016 the Chair in OER had a presentation of AI and machine learning technologies to UNESCO ED and CI colleagues working in or interested in this field.

Result:  From 18 – 20 September 2017, the Slovenian Government will organize together with UNESCO the 2nd World Congress on Open Educational Resources and K4A will run an event on AI for Global education.

 

Goal 4: Ensure inclusive and quality education for all and promote lifelong learning
Goal 4: Ensure inclusive and quality education for all and promote lifelong learning

Background: AI and machine learning (ML), which include technologies such as deep learning, neural networks and natural-language processing, can also encompass more advanced systems that analyse, learn, predict, adapt and potentially operate autonomously. Systems can learn and change future behaviour, leading to the creation of more intelligent devices and programs. The combination of extensive parallel processing power, advanced algorithms and massive data sets to feed the algorithms has unleashed this new era. In the same way as AI is impacting today many sectors (Health, Security, Autonomous vehicles, smart cities, etc.), we believe it will have a profound impact on education. But for this impact to be of benefit to all, everywhere, a collective effort has to be made by all actors: UNESCO has the understanding of where the problems lay and what are the challenges. UNESCO specialists also have a global understanding of the links between Education issues and the different SDGs and therefore the impact that progress in Education can have over the general UNESCO agenda. Scientists have the understanding of the algorithms, technologies which are being deployed today on smaller scales, in experimental settings or on altogether different problems, finaly the industry has the capacity of deploying solutions.

Opportunity: In addressing the SDG 4 on education, it is possible to use AI and machine-learning techniques to model users and fuse large amounts of educational and OER data to enable OERs five characteristics of openness — that is, users can retain, reuse, revise, remix and redistribute the content freely. Organizations seeking to drive digital innovation with this trend should evaluate a number of educational scenarios in which AI and machine learning could drive clear and specific education value and consider experimenting with one or two or more high-impact scenarios. The measure of success for AI applications is the value they create for human lives. In that light, they should be designed to enable people to understand AI systems successfully, participate in their use, and build their trust.

Value proposition: K4A will organise a conference featuring talks from UNESCO Chairs and experts, and leading AI, Big Data and machine learning scientists both from Academia and Industry. These talks would be preceded by online discussions with starting point problems identified by UNESCO Chairs, National Commissions and Personnel in Paris related to UN SDG 4, and scientific discussions aiming at proposing how AI and machine learning technologies may contribute to answering the proposed problems.

Organisation: A 20-person committee will be created, with 10 AI/ML/Big Data scientists (UNESCO compliant in gender issues and all Regions), and 10 UNESCO Chairs, National Commissions personnel or UNESCO Paris personnel. The UNESCO Chairs would provide the challenges, for example “Maintaining the culture, language, history, knowledge for small communities is very difficult. This is often too expensive and there are never the same resources for education as when one wants to teach using one of the more important languages. A solution would be to make it easier to build local material, share it, have the community enrichen it. But there never seems to be the money, the technical skills, the infrastructure to do this. Can you help us?”
The above might be the statement from a specific UNESCO Chair, National Commission or Personnel in Paris who may also provide us with a description of the data available as the numbers reflecting the situation, etc. When these 10 problems have been written out (or at least partially described), the scientists assembled by K4A would discuss and come up with meaningful examples and solutions. The conference would aim to give these discussions a public face. It may also be possible to hold the Ljubljana event in coordination with industry: the question would be posed by the UNESCO experts, the ML scientists would argue what would be feasible, and the companies could attempt to build viable financial models.

Target audience: K4A is in a strong position to link all these actors together: Big-Data/Machine Learning specialists, Industry, UNESCO Chairs, general audience. It has the contacts in the different communities concerned, having been involved with these questions in the past. Their Trustees are Machine Learning specialists, UNESCO Chairs, and are used to organizing large events. Finally, one of their trustees, is currently in charge of the Eric & Wendy Schmidt Data Science for Social Good Summer Fellowship at the University of Chicago (link).

End result: Artificial intelligence, machine learning, and smart things are cited in every major report as technology trends from 2017-2020 and promise an intelligent future. AI and machine learning have reached a critical tipping point and will increasingly augment and extend virtually every technology enabled service, thing or application. The event would result in a “Global Roadmap for solving SDG 4 via the power of AI”.

Proposed timeline:

  • December 2016: make the decision of organizing the event
  • January 2017: recruiting the different committee members – goal is Feb 28th
  • March 2017: the UNESCO Chairs, National Commissions personnel or UNESCO Paris personnel, with the help of K4A, identify the challenges based on SDG 4
  • April 2017: The scientists discuss with the Chairs so that one scientist is in charge of each problem (but it is agreed that discussions are open: other scientists can be invited).
  • May-June 2017: Problems are addressed by the scientists
  • July-August 2017: the different proposals are discussed and K4A people decide which proposals will be fully presented at the conference.

Roles:
K4A personnel with Trustees:

  • Running the conference
  • In charge of publicity (includes the website)
  • Recruiting the scientific experts
  • Interfacing with the UNESCO experts

UNESCO experts are expected to:

  • Meet a very tight timetable
  • Propose (one page) a problem relevant to their chair, area and also to an SDG.
  • Discuss this problem with the K4A person in order to create a two-page document with possible appendices (including links to papers, data, extra analyses) [March 2017]
  • Accept that this document is openly accessible and shared on the K4A webpage
  • Discuss with the scientific expert chosen to champion this cause, and possibly with his assistants
  • Participate in the debriefing which will take place in Slovenia in September 2017

Scientific experts are expected to:

  • Meet a very tight timetable
  • Choose (possibly with a discussion organised by K4A) from 10 two-page documents the SDG problem he/she wants to champion.
  • Discuss with the UNESCO expert to better understand the issues, the possible data, and what would have the most impact, etc.
  • Recruit more assistants he/she will be mentoring