Archive for the ‘AI’ Category

What is Machine Learning

January 20th, 2021 by Graham Attwell

What is machine learning header

I am copying this from Stephen Downes’ ever informative OLDaily newsletter digest. It features an article entitled What is machine learning? – A beginner’s guide posted on the FutureLearn website.

This is quite a good introduction to machine learning. If you don’t know what it is and would like a quick no-nonsense introduction, this is it. Machine learning is depicted “as the science of getting computers to learn automatically.” It’s a type of artificial intelligence, which means essentially that they are software systems that “operate in an intentional, intelligent, and adaptive manner.” The third point is the most important, because it means they can change their programming based on experience and changing circumstances. The article talks about some types of machine learning systems and outlines some application in the field. It’s FutureLearn, so at the end it recommends some course tracks for people interested in making this a career, and just to dangle a carrot, the web page lets you know the median base salary and number of job opening for the program in question.

AI and Edge computing

January 7th, 2021 by Graham Attwell
ball, abstract, pattern

geralt (CC0), Pixabay

A recent MIT Technology Review Insights reports on a survey of 301 business and technology leaders around their use and future planned us of Artificial Intelligence. The survey confirms that the deployment of AI is increasing, not only in large companies but also in SMEs. It also points to the emergence of what is known as edge  comput9ing, using a variety of devices closer to the applied use than cloud computing allows and capable of near real time processing.

38% report of those surveyed report their AI investment plans are unchanged as a result of the pandemic, and 32% indicate the crisis has accelerated their plans. The percentages of unchanged and revved-up AI plans are greater at organizations that had an AI strategy already in place.

AI is not a new addition to the corporate technology arsenal: 62% of survey respondents are using AI technologies. Respondents from larger organizations (those with more than $500 million in annual revenue) have, at nearly 80%, higher deployment rates. Small organizations (with less than $5 million in revenue) are at 58%, slightly below the average.

Cloud-based AI also allows organizations to operate in an ecosystem of collaborators that includes application developers, analytics companies, and customers themselves.

But while the cloud provides significant AI-fueled advantages for organizations, an increasing number of applications have to make use of the infrastructural capabilities of the “edge,” the intermediary computing layer between the cloud and the devices that need computational power.

Asked to rank the opportunities that AI provides them, respondents identify AI-enabled insight as the most important (see Figure 2). Real-time decision-making is the biggest opportunity, regardless of an organization’s size: AI’s use in fast, effective decision-making is the top-ranked priority for large and small organizations.

For small ones, though, it is tied to the need to use AI as a competitive differentiator.

Again, the need for real-time data or predictive tools is a requirement that could drive demand for edge-based AI resources.

Survey respondents indicate that AI is being used to enhance current and future performance and operational efficiencies: research and development is, by a large margin, the most common current use for AI, used by 53% of respondents, integrating AI-based analytics into their product and service development processes. Anomaly detection and cybersecurity are the next-most-deployed AI applications.

Large organizations have additional priorities: 54% report heavy use of robotic process automation to streamline business processes traditionally done by humans, and 41% use AI in sales and business forecasting. For organizations with AI strategies, 40% rely on robotic process automation, and 42% use AI to estimate future sales.

MOOCs and Artificial Intelligence – Potentials for the Professional Development of VET Teachers and Trainers

December 21st, 2020 by Graham Attwell
processor, cpu, computer

ColiN00B (CC0), Pixabay

It does not seem likely that we are going to participate in any face to face conferences in the near future. But conferences are continuing online and in some ways there are increased opportunities for sharing ideas and knowledge. Anyway that was a preamble for the latest abstract that I, together with Sohia Roppertz and Ludger Dietmer have submitted for the 4th Crossing Boundaries on Vocational Education and Training in 2021.

1        Introduction

The growing use of Artificial intelligence (AI) and other technological innovations are leading to a fundamental change in the world of work, as tasks previously performed by humans can now potentially be performed/assisted by computers and computer-controlled machines (Brynjolfsson & McAffee, 2014; Dengler & Matthes ,2018).

This digital transformation places VET under high pressure to adapt (Seufert, 2018), to provide professional action competence within non academic technical, social, commercial and other occupations. VET schools and their teachers and trainers have the central task of preparing learners for the changing world of work.  Technological change also is impacting the organisation of  VET schools through the introduction of big data and e-government and at the implementation level in connection with e.g. adaptive learning systems and learning analytics (Seufert 2018).

Against this background, the question arises how vocational school teachers and trainers can be prepared for these tasks. While a survey undertaken through the Taccle AI project found most vocational teachers and trainers recognised the importance of AI, there are presently few opportunities for professional development (Author, forthcoming).

This paper reports on work in progress through the EU Erasmus Plus funded Taccle AI project examining the impact of AI on Vocational Education and Training, led by the University of Bremen Following initial research and the development of a Resource Toolkit, the project is developing a Massive Open Online Course (MOOC), to be offered in English and  Germany and expected to be launched May 2021.

In section two of the paper we will examine the vision behind MOOCs and in section three will look at the different dimensions of a MOOC for professional development for teachers and trainers around AI. This will be followed by a discussion of how cooperative and project related MOOC design can support teachers and trainers in developing their professional practice. Finally we will draw some initial conclusions based on our work.

  1.       Theoretical background

The first MOOC, led by George Siemens and Stephen Downes was in 2008 around the topic of Connectivism. According to Downes (2012) it was based on the realization that the use of distributed open resources would support – with ease – an attendance in the thousands. The vision grew out of the idea of Open Education, where everybody could access free online courses. The idea quickly took off, especially with the launch of Coursera and Udacity. Although the founders of these companies saw their innovation as disruptive to traditional education institutions, universities have been quick to pick up on the potential of MOOCs. In Europe one of the biggest MOOC providers is OpenLearn, through the UK Open University leading a consortium of educational providers.

2.1       Vision behind MOOCs

There has been and continues to be discussion over pedagogic approach to MOOC design, with advocates of so called cMoocs emphasising the active contribution of participants, using digital platforms and technologies while so called xMOOCs, for example from Stanford University, are more focused on the transmission of knowledge.

MOOCs are increasingly being used for professional development, for instance by companies like Siemens, and for teachers and trainers.

  1. MOOCs and AI – two dimensions

There are two key dimensions to the AI MOOC.

3.1       MOOCs as a Way to Learn about AI

The first is MOOCs as a way to learn about AI. This in turn has three key foci. After an initial introduction the course will examine both the impact of AI on the Changing world of Work and its implications in terms of skills, tasks and consequently curriculum. The second focus will be on the use of AI for teaching and learning in VET. The final section will examine the ethical implications of AI for VET.

3.2       Artificial Intelligence powered MOOCs

The second key dimension will be the integration of AI into the MOOC platform. While this work is still under development it is intended to incorporate Natural Language processing for the production of materials and Learning Analytics within the MOOC Platform.

  1.             Implementation of cooperative and project related (more interactive) MOOCs design in professional development of VET teachers and trainers

New forms of learning are needed for AI in VET. MOOCs including practice and project-based learning can be used both in VET courses and for training teachers and trainers. The paper will discuss how MOOCs can be designed through a new didactical approach to teaching and learning.

The article will explore the concept of additional qualifications within apprenticeship training and how such concepts can be adapted to different European vocational settings. It will show how the new arrangements affect the teaching and learning of  VET students (e.g. mechatronic apprentices), and new roles for VET teachers and trainers. Discussion will be based around examples, for instance the mechatronic students are working and learning in school based and training factory labs themselves developing AI projects (Author et Al, 2020a). This functional learning material allows student teams to plan, prepare, realize and demonstrate projects in which for example autonomous driving or robot functions are programmed, tested and presented to bigger audiences. These practice based developments will be incorporated in the MOOC.

  1. Conclusions

This paper will examine the question of whether MOOCs are suitable for the continuing education of VET teachers and trainers in the context of Artificial Intelligence. It will also clarify what kind of MOOCs types and settings can be connected to projects, how MOOCs can be implemented by VET teachers and trainers and how they have to be structurally designed in didactical terms. This includes the question how teachers and trainers have to be prepared for such new learning arrangements in order to develop vocationally oriented MOOCss (Author at Al, 2020b).

References

Author, Bekiaridis, G., Author., Tutlys, T., Perini, M., Roppertz, S., & Tutlys, V. (2020a). Artificial Intelligence in Policies, Processes and Practices of Vocational Education and Training. Institut Technik und Bildung. https://doi.org/10.26092/elib/307

Author, Author., Tutlys, T., Author, & Perini, Marco. (2020b). Digitalisation, artificial intelligence and vocational occupations and skills: What are the needs for training teachers and trainers? In C. Nägele, B. E. Stalder, & N. Kersh (Eds.), Trends in vocational education and training research, Vol. III. Proceedings of the European Conference on Educational Research (ECER), Vocational Education and Training Network (VETNET) (pp. 30–42). https://doi.org/10.5281/zenodo.4005713

Downes, S. 2012, The Rise of MOOCs, https://www.downes.ca/cgi-bin/page.cgi?post=57911, accessed 15 December, 2020

Brynjolfsson, E., & McAfee, A. (2014). The second machine age: Work, progress, and prosperity in a time of brilliant technologies. WW Norton & Company.

Dengler, K. & Matthes, B. (2018). Substituierbarkeitspotenzial von Berufen. Wenige Berufsbilder halten mit der Digitalisierung Schritt. IAB-Kurzbericht.

Seufert, S. (2018). Flexibilisierung der Berufsbildung im Kontext fortschreitender Digitalisierung. Bericht im Auftrag des Staatssekretariats für Bildung, Forschung und Innovation SBFI im Rahmen des Projekts «Berufsbildung 2030 – Vision und Strategische Leitlinien» Zugriff unter: https://edudoc.ch/record/132323 (1.12.2020)

Author, (forthcoming). Artificial Intelligence and Vocational Education and Training – Perspective of German VET teachers. Proceedings of the European Distance and E-Learning Network Research Workshop, 2020 Lisboa, 21 – 23 October 2020.)

Keeping up with the data

December 14th, 2020 by Graham Attwell

AI as I am sure you all know by now, is largely reliant on the black gold, data, and preferably lots of it. With the speed that things are developing at, its pretty hard to keep up with the world of AI and data. In the past I subscribed to the excellent free weekly MIT technology Review, but sadly that went the way of paid for subscription services.

But the London and Leeds based Open Data Institute continues to publish a free weekly newsletter, The Week in Data. And it tends to favour more a social approach than some other more technology heavy news services. Here is an excerpt from last Friday’s issue’

If you’re looking to stream an anxiety-inducing film about the rise of robots over your Christmas break, you may not think to look in the documentary category. But iHuman, a gloomy (and seemingly slightly dramatised) take on the world of AI, is released this week in cinemas and (and online), and sits firmly in the ‘documentary’ genre. Film-maker Tonje Hessen Schei speaks to a range of interviewees, including Elon Musk’s computer scientist, to explore what will happen ‘when robots become smarter than humans’. It has been described by The Guardian as ‘an eye-opening film if your anxiety levels are up to it’. We’d love to hear – once you’ve had a look – whether it filled you with joy, dread or a painful amount of eye-rolling.

You can subscribe to The Week in Data here.

Does only AI make a good university?

December 10th, 2020 by Graham Attwell
town musicians of bremen, cat, dog

Clker-Free-Vector-Images (CC0), Pixabay

Plans for a dedicated AI university campus in Bremen are causing a controversy. The plan is backed by a consortium  of German software giant SAP, Chinese IT firm Neusoft and the German Research Center for Artificial Intelligence.

It is intended to take over the Jacobs University, a private university in the north of the city founded in 1999. The university has been funded up to now by the food company Jacobs, but they have announced they are ceasing their financial support. The university has always been controversial with questions over the need for a private university competing withe the state funded Bremen University.

THE reports that the Bremen government hopes that a dedicated AI campus will boost the local economy, attracting skilled workers and companies, although the city and firms involved have not yet drawn up detailed plans.

But there are fears from academics at Jacobs that the university, which sells itself on its interdisciplinary campus, will end up with too narrow a focus.

“AI is not necessarily a bad thing,” said Professor Bau. “The big question is: only AI, does that actually make a good university? Universities and university education are something that does not have a small focus.”

Developing Competences for the AI Era

December 1st, 2020 by Graham Attwell

UNESCO are increasingly active in the development in the use of Artificial Intelligence in Education. Next week – 7 and 8 of December they are hosting an International Forum on AI and the Futures of Education (AIED). The theme of the Forum is  Developing Competencies for the AI Era. the Forum will bring togther education and technology experts from around the world to discuss AI skills for the futures of education and AI as a common good for education. UNESCO say participants will share policies and practices in defining the competencies required in the AI era, and examine strategies to prepare all people to live and work with AI effectively.

“The concept of futures in the plural is used to recognize that there is a rich diversity of ways of knowing and being around the world. The plural form also acknowledges that there are multiple dimensions to the future and that there will likely be various desirable and undesirable futures – all of which will vary greatly depending on who you are and where you stand. Rather than attempting to chart a single future, looking at futures in the plural validates multiple possible and desirable futures of humanity on our shared planet.

The Futures of Education initiative aims to generate discussion and action on the role of education, knowledge and learning in view of the predicted, possible and preferred futures. Such re-visioning of knowledge, education and learning is more relevant than ever. Accelerated technology transformations over recent years, in particular in the field of Artificial Intelligence, and their rapid deployment in work, life and learning have profound implications for the futures of education. The Forum will provide an opportunity to discuss these implications and the transformative potential of AI on education.”

Attendance is free and participants can register at https://aiedforum.org/#/home.

FutureLearn team up with Microsoft for online AI course

November 18th, 2020 by Graham Attwell

As many of you will know, FutureLearn is the UK Open Universities MOOC arm, run in conjunction with an international consortium of universities. But, I guess like everyone else, FutureLearn is under pressure to make some money. Their first go was offering paid for certificates for course completion. Another attempt has been to persuade people to sign up for an annual subscription, keeping courses open for a year if they pay.

The latest is to partner with industries for courses providing micro accreditation, in some cases industry recognised. So in December Future Learning is launching “Artificial Intelligence on Microsoft Azure: Machine Learning and Python Basics‘, created by CloudSwft and inc conjunction with Microsoft. “On this microcredential”, they say ” you’ll address this challenge by developing key AI skills that can serve as the first steps towards becoming an AI engineer, business analyst, or AI professional.” And, “Yes. If you successfully complete this microcredential, you’ll receive a voucher to sit a separate exam to earn the Microsoft Azure AI Fundamentals (AI-900) and Microsoft Azure AI Engineer Associate (AI-100) certification.”

Why would FutureLearn be giving away vouchers for sitting Microsoft exams. It could be because the 15 week course costs 584 Euros to enroll.  Much as I like microcredentially, this seems a long way from FutureLearn’s past MOOCs free for participation. And if as the course information claims, “artificial intelligence skills are frequently listed among the most in-demand workplace skills in the current and future job market, as organisations seek to harness AI to revolutionise their operations” and “employers are faced with a shortfall of qualified candidates” surely this is an area where public education and trainings services should be providing online course, rather than restricting access to those who can afford to pay for learning new skills.

 

Workshop on Ai and Vocational Education as part of European Vocational Skills Week

November 9th, 2020 by Graham Attwell

geralt (CC0), Pixabay

This week is European Vocational Skills Week.

And as a partner of the European Vocational Skills Week the Taccle AI project, is organising an online workshop on “Artificial Intelligence for and in VET” on Tuesday 10 November 15:00 – 16:30 CET. Our Taccle AI project partners from five European countries will welcome you.

About the Workshop:

AI is particularly important for vocational education and training (VET) as it promises profound changes in employment and work tasks. Not only are some jobs vulnerable and new jobs likely to be created but there will be changing tasks and roles within jobs, requiring changes in initial and continuing training, for those in work as well as those seeking employment. This will require changes in existing VET content, new programmes such as the design of AI systems in different sectors, and adaptation to new ways of cooperative work with AI.

For VET teachers and trainers there are many possible uses of AI including new opportunities for adapting learning content based on student’s needs, new processes for assessment, analysing possible bottlenecks in learners’ domain understanding and improvement in guidance for learners

In our workshop we will explore these issues with short inputs and breakout sessions for discussion by participants around key issues.

Register now (here)!

Finishing the last reports for TACCLE4 CPD project – Handing over the torch for other runners

October 31st, 2020 by Pekka Kamarainen

During the last few months I have surprised myself by producing three new reports for the ongoing EU-funded TACCLE4 CPD project. As regular readers of this blog will know, the project has been working with strategies for promoting digital competences of teachers and trainers in different educational sectors. And, as a contrast to the three earlier TACCLE projects, the fourth one had the task to shape models and concepts for continuing professional development in educational establishments and training organisations. Furthermore, my role in the project has been to address the task for the field of vocational education and training (VET) and to bring into the project the legacy of the Learning Layers project (in which I had been working for many years).

Now, the project is in its final phase and at the same time my career as an active researchers has come to an end. During the last few weeks I have surprised myself and others by producing three new reports for the project – in addition to the five ones that I had produced by the end of last year. Below, I want to make some comments on these newer reports and how they enrich the group picture of the earlier VET-related reports.

Report 5: Promoting digital competences beyond the accustomed realm of ICT skills

During the project I had been writing blogs on innovative activities to promote digital competences via civic learning and via introduction of specific applications of the Learning Toolbox (the main product developed in the Learning Layers project). To me, the Finnish idea of developing a broad-based introductory course on artificial intelligence (AI) and its impact for the entire civil society was very inspiring. Equally, the recent progress in using the Learning Toolbox (LTB) at different contexts was impressive. In the initial pilot context, vocational trainers made new efforts to support vocational learning during the lockdown. and after the closure period they added new features to using LTB in apprentice training.

Parallel to this, other applications of LTB – mainly the use  of LTB to shape ePosters for conferences (that are shifting from face-to-face events into online events) is becoming widespread. Here, it is worthwhile to note the spread from regular conferences into other kinds of of online events – such as multiplier events of EU-funded projects. Furthermore, I got informed of the progress with LTB showcases. Firstly they had been shaped for particular conferences to give a group picture of the ePosters for the respective conference. In a more mature phase the LTB showcase was used to give a group picture of all conferences and online events that were working with ePosters – of which some exemplars were portrayed in this ‘all stars’ showcase. I was happy to give visibility to the start-up company Kubify (founded by the former LTB-developers) due to their latest achievements. Altogether, this report demonstrated the progress of promoting digital competences beyond the accustomed realms of ICT skills.

Activity Report on the German Multiplier Event at the Training Centre Bau-ABC Rostrup

During my working visit in Bremen (after a long break) I had a chance to organise a Multiplier Event that took place as face-to-face event in the training centre Bau-ABC Rostrup (with whom we had worked intensively in the Learning Layers project). For me it was a pleasant opportunity to meet several trainers of the centre – some of which had been our counterparts for years, whilst some were newcomers. In my input I gave an overview, how the co-design processes (that led to to the shaping of the LTB) and the parallel training activities (that were completed with training based on Theme Rooms) worked hand in hand in the Learning Layers project. Then I gave some insights into the idea of different innovation paths (for introducing digital competences in the field of VET) and how they can be addressed in a revisited framework for Theme Room Training 2020.  In addition to this I presented the new TACCLE4 CPD Showcase that pulled together the VET-related work in the Learning Layers project and in the current project.

It was interesting to note that the discussion moved on from my inputs to the practical challenges (regarding the use of LTB as support for training) and to the possibilities to take further steps with their internal training. So, in this respect the session became a genuine multiplier event. Indeed, it was concluded with a commitment to start a new round of theme room sessions – based on the ideas and needs of the active trainers and with focus on improving the use of LTB in their context.

Report 2b: The TACCLE4 CPD Showcase as a new collection of online resources

My final report for the project TACCLE4 CPD presented the above-mentioned LTB showcase for this project and how it came into being. As the pre-history I presented the earlier pilot activity to introduce the work with ePosters into the European Conference on Educational Research (ECER) in the year 2018 in Bolzano. This was a limited pilot project with some ePosters for the network program of VETNET (the European Research Network for Vocational Education and Training). As a further step from this phase the report introduced an overarching showcase for the TACCLE4 CPD project. This showcase provided on the one hand comprehensive LTB-stacks thet presented all VET-related reports for the Learning Layers project and for the TACCLE4 CPD project. Then, it introduced thematic stacks that focused ot the use of Open Educational Resources (OER) in the field of VET and outlined the Theme Room Training 2020 framework.

Altogther, this report gave a picture of a new kind of an online resource environment. At the same time it invited the users to consider their own innovation paths and to think of their own ways to shape training with Theme Rooms.

I guess this is enough of the latest reports and of the messages I want to pass at the final phase of the project. Now that I am finishing my career as an active researcher, I have the feeling of being a runner with the torch of the Olympic Games. I am coming to the point in which I have to had out the torch for new runners – the ones in research, the ones in the software development and the ones developing their training approaches. My message to them is the following: You need to take the fire from the past, not the ashes. And: You need not go back to the long and winding roads of your predecessors – you need to find each other in the present date circumstances.

More blogs to come … (but from a different perspective)

Data Driven Science

September 29th, 2020 by Graham Attwell

This diagram is from a tweet by  Data Driven Science (@DrivenScience).

Artificial Intelligence they say, is the broad discipline of creating intelligent machines.

Machine Learning refers to systems that can learn from experience.

Deep Learning refers to experience on large data sets.

  • Search Pontydysgu.org

    Social Media




    News Bites

    Racial bias in algorithms

    From the UK Open Data Institute’s Week in Data newsletter

    This week, Twitter apologised for racial bias within its image-cropping algorithm. The feature is designed to automatically crop images to highlight focal points – including faces. But, Twitter users discovered that, in practice, white faces were focused on, and black faces were cropped out. And, Twitter isn’t the only platform struggling with its algorithm – YouTube has also announced plans to bring back higher levels of human moderation for removing content, after its AI-centred approach resulted in over-censorship, with videos being removed at far higher rates than with human moderators.


    Gap between rich and poor university students widest for 12 years

    Via The Canary.

    The gap between poor students and their more affluent peers attending university has widened to its largest point for 12 years, according to data published by the Department for Education (DfE).

    Better-off pupils are significantly more likely to go to university than their more disadvantaged peers. And the gap between the two groups – 18.8 percentage points – is the widest it’s been since 2006/07.

    The latest statistics show that 26.3% of pupils eligible for FSMs went on to university in 2018/19, compared with 45.1% of those who did not receive free meals. Only 12.7% of white British males who were eligible for FSMs went to university by the age of 19. The progression rate has fallen slightly for the first time since 2011/12, according to the DfE analysis.


    Quality Training

    From Raconteur. A recent report by global learning consultancy Kineo examined the learning intentions of 8,000 employees across 13 different industries. It found a huge gap between the quality of training offered and the needs of employees. Of those surveyed, 85 per cent said they , with only 16 per cent of employees finding the learning programmes offered by their employers effective.


    News from 1994

    This is from a Tweet. In 1994 Stephen Heppell wrote in something called SCET” “Teachers are fundamental to this. They are professionals of considerable calibre. They are skilled at observing their students’ capability and progressing it. They are creative and imaginative but the curriculum must give them space and opportunity to explore the new potential for learning that technology offers.” Nothing changes!


    Other Pontydysgu Spaces

    • Pontydysgu on the Web

      pbwiki
      Our Wikispace for teaching and learning
      Sounds of the Bazaar Radio LIVE
      Join our Sounds of the Bazaar Facebook goup. Just click on the logo above.

      We will be at Online Educa Berlin 2015. See the info above. The stream URL to play in your application is Stream URL or go to our new stream webpage here SoB Stream Page.

  • Twitter

  • Why do so many professional, middle-class Brits insist they're working class? | Sam Friedman theguardian.com/commentisfree…

    About 4 days ago from Cristina Costa's Twitter via Twitter Web App

  • Recent Posts

  • Archives

  • Meta

  • Categories